Room 303A-339A
Hubert Humphrey Building
200
Independence Avenue, S.W.
Washington, D.C.
Barbara Starfield, M.D., Chair
Jeffrey S. Blair, M.B.A.
Simon
P. Cohn, M.D., M.P.H., FACP
Kathleen A. Frawley, J.D., M.S., RRA
Kathleen Fyffe, M.H.A.
John R. Lumpkin, M.D., M.P.H.
Clement
Joseph McDonald, M.D.
William Braithwaite, M.D., Ph.D.
Judy Ball
Marjorie S.
Greenberg
Robert Moore
Gail R. Janes, Ph.D., M.S.
Opening Remarks and Introductions
Payers' Perspectives on Claims Attachments
State/Public Health Perspectives
Providers' Perspectives on Claims Attachments
Claims Attachments Project: Steven Barr
DR. COHN: Could I ask everybody to be seated, please. I am Dr. Simon Cohn with Kaiser Permanente.
I want to welcome everyone to the hearings on health claims attachments for the Subcommittee on Health Data Needs, Standards and Security of the National Committee on Vital and Health Statistics.
I am, obviously, acting chair. I think we will be seeing some other members of the committee arriving, hopefully, in the next couple of minutes, but I wanted to get everything started.
As is the usual for this committee, why don't we start by introducing members of the committee, followed by members -- participants on our first panel and then members of the audience.
Jeff, would you like to start?
MR. BLAIR: I am Jeff Blair with the Medical Records Institute.
DR. FRAWLEY: I am Kathleen Frawley of the American Health Information Management Association.
DR. BRAITHWAITE: Bill Braithwaite, staff to the subcommittee.
MS. BALL: Judy Ball, also staff to the subcommittee.
MS. KORMANEC: Rosemarie Kormanec with Principal Health Care, Incorporated.
DR. MUSCALUS: Dr. Rob Muscalus with Xact Medicare Services.
MR. ANDERSON: Pete Anderson with United HealthCare Corporation.
[Introductions off microphone.]
DR. COHN: Is that everyone?
Marjorie, would you like to introduce yourself?
MS. GREENBERG: Sorry. Marjorie Greenberg, executive secretary to the committee.
DR. COHN: Well, as the acting chair, I basically want to go through the agenda for the day and just make a comment or two about this meeting.
I think you all have a copy of the revised agenda. We will start out -- we are going through a series of hearings and panels on issues related to claims attachments, starting from the payers' perspective this morning, from 9:15 to about 10:30. From 10:30 to 11:00, we will take a break and then continue on with other payer perspectives and state and public health perspectives.
This afternoon, we will talk about provider perspectives on claims attachments and then at the end of the afternoon, Steve Barr from HCFA will be talking about the Claims Attachments Project that he has been involved with, I think, for nearly a year now.
Now, I do want to comment that I think, at least to me as a physician, I think that these are among the more important hearings that we are going to be holding around administrative standards. In many ways, claims attachments are beginning to get to some of the more interesting issues around administrative transaction standards and begin to get close to the issues of clinical standards and all of this.
So, I am delighted that you are here. Certainly, I don't think this will be -- we will not have solved the problem by the end of tomorrow, but I think we will have made some good headway. So, I want to thank you all for coming.
Now, before the panel starts, are there any comments? Bill, would you like to make any comments this morning?
DR. BRAITHWAITE: No, just to note the structure of the hearings is -- because this is an area where there is no active national ANSI standard available for the Secretary to adopt, that the hearing is taking the tact of hearing, first of all, from the payers, who are the people who need the information contained in claims attachments and then to hear from the providers or the people who have to provide that information to satisfy the payers and then to talk -- well, not only the payers need this information, but state and public health people need the information.
So, we are going to hear from them this morning as well and then from the providers who have to hear it, have to provide the information and then from Steve Barr, who is the project leader of the attempt of a couple of standard committees to actually put together a national ANSI standard to satisfy this need for electronic transmission of claims attachments.
Tomorrow, we will hear from vendors, who will be responsible essentially for providing the software to do this and then from a variety of content experts, who may be called upon to be the committee that decides and negotiates for the industry what the content of the actual medical messages will be and from the SDOs that will be negotiating the mechanism for getting the message from one place to another. Then we will have the afternoon tomorrow for the subcommittee to discuss what we plan to do from here, having had all of that input.
DR. COHN: I would also like to introduce Kathleen Fyffe, another member of the committee, who just arrived.
Any other comments of the committee before we launch into our first panel? No. Okay.
Peter, would you like to lead off the panel discussion?
MR. ANDERSON: Mr. Chairman, members of the subcommittee, I want to thank you for the opportunity to appear at the hearings. My name is Pete Anderson. I am an information systems director at United HealthCare Corporation. United HealthCare provides a full spectrum of resources and services to help people achieve improved health and well-being through all stages of life.
Over 75,000 employers off our products to millions of Americans. I am responsible for electronic data interchange in our government operations. I have 30 years of technical and management experience in information systems, including 17 years with the electronic exchange of Medicare claim data.
I am the United HealthCare representative to X12 and I worked closely with HCFA to develop the implementation guides for the Part B Medicare electronic claim and remittance transactions.
On behalf of the management of United HealthCare, I am happy to be a part of this session and to assist in any way that I can.
I would now like to respond to the questions posed by the subcommittee.
Our definition of "claim" versus "attachment": A claim is a request for payment of services provided. This is usually a HCFA-1500, UB-92 or the electronic equivalent. A claim attachment is extra documentation that supports or supplements the request for payment. A provider might submit an attachment to substantiate higher reimbursement or we might request an attachment to make a determination regarding coverage or medical necessity.
Supporting information can also be requested after a claim has been received by a payer. For purposes of defining "electronic claim attachments," we prefer to include these items as well, even though they are not attached to the submitted claim.
Second, differentiating claim versus attachment information: The claim should provide demographic information for the provider and member, policy information, dates of service, charges, diagnosis and procedures. In most cases, this is all that is necessary to process the claim to payment.
When required, claim attachments should support the medical necessity and reasons for procedures rendered, past and future frequencies and duration of treatment and outcomes of treatment. Claim attachment information should be specific to the type of attachment and include more detail on the type of services provided.
We believe a key objective of any approach to differentiation should be to simplify the implementation of both standards for as many submitters as possible.
Types of claim attachments: We request the following types of claims attachments: operative notes, pathology reports, treatment plans, medical necessity reports, progress notes, consultation reports, additional ambulance information, procedure reports, medical history.
We request these items using form letters, telephone calls and explanations of benefits. We receive this information on paper.
We request the following other types of documentation: explanation of benefits and automobile liability information for primary payers; tax information, such as W-9 forms; verification of dosage for injections; verification of date of death; verification of health insurance claim numbers; verification of diagnoses.
We also request these items using form letters, telephone calls and explanations of benefits and we also receive this information on paper.
Fifth question on attachment purpose: These attachments and documentation provide the claims processor or medical reviewer with information to determine coverage, medical necessity and which payer is primary. This is needed to determine the benefit due. Some information is also used to check for fraud and abuse.
What would be aided by standard electronic exchange? With standardization, providers could expect to be asked the same questions in the same way by all payers. Providers would submit the same information in response to the standard requests. This would make the process more consistent for both providers and payers.
Both the payer request for additional information and the provider response can be handled electronically. Electronic request and return of additional information would speed up the payment process and generate savings for postage and paper handling. If the additional information were available in the provider's system, clerical retrieval could also be eliminated.
In 1996 and 1997, United HealthCare participated in a proof of concept with HCFA. We converted several of our computer-generated form letters into electronic requests for additional information.
From this pilot, we learned it is feasible to ask for such supplemental information electronically. In fact, the participating providers suggested the next logical step was for them to return the requested information electronically.
What is not conducive to standardization? While the process will benefit from standardization, there are some aspects, which may gain less immediate benefit. Some attachments, such as treatment plans and operative notes, are non-codified text. While these may be electronically delivered to the appropriate reviewer more rapidly, they will still require human interpretation.
Also, some attachments prove that services were actually rendered or that they were ordered by a physician. For example, physician certification is required for many therapy services. Since this certification is indicated by a physician's signature, an electronic attachment may be considered insufficient.
The relationship between attachments and the medical record: Claim attachments are not typically part of the medical record but are an extract or summary of specific information contained in the record.
The entire medical record: We request the entire medical record in only a few circumstances, such as to review extended lengths of stay for psychiatric or rehabilitation treatments; to investigate preexisting conditions; to check on procedures that appear unrelated to the submitted diagnosis and, after payment, to substantiate certain appeals or to investigate potential fraud and abuse.
Approving the content of claims attachments: We believe that the National Uniform Billing Committee and the National Uniform Claim Committee should have responsibility for approving the content of claims attachments.
For attachments dealing with basic non-skilled services, this would be sufficient. For other attachments, additional collaboration would be appropriate.
Attachments for complex skilled services should be reviewed before they are released to the NUCC or NUBC for approval. For example, attachments for respiratory therapy might be reviewed by the Health Care Financing Administration in collaboration with the National Association of Respiratory Therapists, Pulmonary Physician Specialists and health plans.
Attachments covering services considered prone to fraud should be reviewed by fraud agencies, such as the FBI and health plan fraud units before they are released to the NUCC or the NUBC for approval.
Protocol for data content: First, we believe that multiple attachment types should be developed, rather than one or two all-inclusive formats. This will reduce the impact on existing implementations when formats must be modified in the future. It will also reduce the amount of data transmitted for each claim and limit it to that which is essential.
Second, the ASC X12 patient information transaction set, the "275," is a standard transaction that can be used to return claim attachment data. Work is progressing on the implementation guide describing how this "envelope" will carry a variety of claim attachments within it. This architecture will accommodate new attachments as they are identified in the future. We believe this concept of a standard envelope to carry attachments is appropriate.
Third, we believe that the standard claims attachment contents for this envelope should be constructed from existing standards, such as Health Level 7, whenever possible. This incremental approach will take advantage of work already done to define supplemental claim information.
And, fourth, since there are many types of attachments, the standard should be explicit and unambiguous about the format to be used for each type. An implementation guide or other specification covering the various attachments must be available to providers and health plans. As new standard attachments are developed, they will need to be added to the specification.
United HealthCare endorses the subcommittee's charter and invites further discussion in any area of concern. Thank you again for this opportunity.
DR. COHN: Do you have a copy of your testimony?
MR. ANDERSON: Yes, I do.
DR. COHN: Our next speaker is Robert Muscalus, please.
Obviously, afterwards, we will have a discussion and take questions from all.
DR. MUSCALUS: Good morning. My name is Dr. Robert Muscalus. I am the vice president of medical affairs and medical director for Xact Medicare Services. I want to thank the members of this committee for the opportunity to discuss the issue of health care claims attachments. Also in attendance with me from Xact are Julie Meehan and Mary Lynn Bushman, who provide support to our electronic data interchange systems.
Xact Medicare Services is the name under which High Mark, Incorporated performs its obligations as a Medicare Part B contractor. High Mark was formed last year as a result of the consolidation of Pennsylvania Blue Shield and Blue Cross of Western Pennsylvania.
High Mark has subsidiaries, which engage in the delivery of various forms of health care products, such as Medicare Part A and Part B, health maintenance organizations, point of service plans, traditional indemnity insurance, in addition to vision and dental products. My comments this morning, however, are based upon my perspective as a medical director for a Medicare Part B carrier.
High Mark, including Xact, has prepared a document in response to the questions prepared by the National Committee on Vital and Health Statistics. Copies are being distributed and additional copies can be made available if requested.
Xact is currently the nation's largest Medicare carrier, servicing approximately 85,000 providers and 3 1/2 million beneficiaries in Pennsylvania, New Jersey, Delaware and the Washington, D.C. metropolitan area. I would like to point out that next month the contract for Delaware and the Washington, D.C. metropolitan area will be transferred to Trail Blazer Health Enterprises.
Each year, Xact processes approximately 73 million claims, resulting in annual payments of approximately 5 1/2 billion dollars. One of my primary roles involves the implementation of program safeguards to protect the fiscal integrity of the Medicare trust funds.
One way this is accomplished is through the development of medical policy. Medicare contractors are required to pay only for services, which are covered by the Medicare program and which are medically necessary, in addition to applying the relevant statutes, regulations and program instructions.
Our medical policies reflect these requirements and we must ensure that our claims processing systems apply the logic defined by each policy. We do not have a precertification or a preauthorization process, which allows for review and payment approval before all services are rendered. Rather, we generally process and pay claims for services, which have been provided and then monitor the utilization of these services in order to detect unusual or aberrant patterns, which may reflect inappropriate utilization or inappropriate billing patterns.
This pay and chase process is costly in terms of both financial and human resources. As a result, carriers attempt, whenever possible, to apply certain logic on a prepay or automated basis to help ensure medical necessity. For example, certain laboratory claims, such as glucose and cholesterol, will pay only if ICD-9 codes are listed, which support the medical necessity of the test.
Unfortunately, this type of automated logic is difficult to apply in all cases. As our published statement indicates, we often require information that is not included on a claim form before a payment determination can be made.
Most of the information we require is in written form and includes items, such as progress notes, operative reports, ambulance trip sheets and other components of the medical record. Other items may also be requested, such as photographs, visual field reports and x-rays.
This information may be requested from the entity who submitted the claim or it may be requested from other sources. For example, physicians may submit claims with certain modifiers to indicate that additional payment is being sought as a result of unusual or difficult circumstances.
In order the process the claim, we may request the operative report from the hospital where the surgery was performed. Of course, when these additional measures are taken, more time is needed to complete the processing of a claim and to reimburse the provider or supplier.
Providing a mechanism to transmit this information electronically will result in greater efficiency, in addition to enhancing timely and accurate claims payment. Utilizing a uniform system will provide some relief to providers who currently are faced daily with numerous requirements, forms and processes from payers in both the public and private sectors.
Payers will benefit by being able to process claims in a timely fashion and as experience with this new technology grows, payers will be able to identify additional areas where such information may be required before payment is made. This will help to not only improve the likelihood that payment is made only for medically necessary services, but it will also help in our ongoing efforts to combat fraud and abuse.
In addition to reviewing information on a prepay basis, information may also be reviewed on a post-pay basis. For example, medical records may be requested from a physician identified as standing out significantly or being aberrant from his or her peers on the basis of the number of specific services billed to the Medicare program.
A review of these records may indicate that all services were appropriate and billed correctly. At the same time, we may discover that the services were not medically necessary, were never actually performed or that the services billed did not accurately reflect the services performed.
Such reviews may result in requests for overpayments, the implementation of provider specific prepay audits or a referral to our fraud unit for further investigation. It is important to identify potential problem areas with an electronic claims attachment process.
Issues, such as patient confidentiality and system security, must be thoroughly researched so that adequate controls can be put in place to maintain the confidence of all parties, especially patients and their families.
I appreciate having the opportunity to comment on this issue and encourage you to continue to solicit the input from organizations, such as ours, as the process proceeds. In addition, I believe the work of the Proof of Concept Team is valuable and that this team should be continue to be actively involved in this project.
I recognize that this is a significant undertaking, which may ultimately impact each and every one of us. However, I am equally confident that by working together, we can overcome obstacles to implement a system, which will benefit us and our families and at the same time help to improve the cost and delivery of health care to all patients.
Thank you.
MS. KORMANEC: Good morning. Madame Chairman, members of the subcommittee, thank you for this opportunity to provide testimony about the utilization and content of health care claims attachments.
My name is Rosemarie Kormanec. I am currently a project manager in the Corporate Health Economics Department of Principal Health Care, Incorporated.
Formed in 1987, Principal is a national health care company with extensive experience in group medical benefits, medical management, plan and benefit design. The company provides a continuum of health care choices and services to over 1.75 million HMO, POS, PPO, Medicare and Medicaid members.
I have been in the health care industry for over 14 years. My areas of expertise include procedural coding, utilization management, case management, HEDIS, information systems, claims, provider reimbursement, data quality, health care data analysis and a clinical nursing background.
I have been involved with claims transactions as the project lead for the implementation and support of a data quality review process. This testimony is based upon my experience in the health care industry and with health care claims and may not in all cases reflect the policies and procedures of Principal Health Care, Incorporated.
My testimony regarding health care claims attachments is also presented on behalf of the American Association of Health Plans, AAHP. AAHP represents over 1,000 HMOs, PPOs and similar plans that provide health care for more than 140 million Americans nationwide. AAHP applauds the administrative simplification requirements of the Health Insurance, Portability and Accountability Act of 1996 and believes that the migration toward uniform standards for electronic business and health care transactions would be an efficient and effective way for transmitting financial and administrative data.
I hope to provide the committee assistance in evaluating the data requirements and options available for the standardization of health care claims attachments, based upon the following responses and industry examples.
Definition of a "claim" versus a "claim attachment": A claim is an electronic or paper record generated after a health professional or facility provides care to a patient. The transaction record includes patient demographics, location of service, rendering provider or facility, services provided, diagnosis and reimbursement information.
A claim attachment is the information that supports the services on a claim. The contents of an attachment provides the cognitive results of an evaluation or encounter supporting laboratory or diagnostic results, operative reports, case management plans, patients' responses to treatment and referrals for services.
How should we differentiate what is appropriate for a claim versus an attachment? The information appropriate for the standard format as a claim is that which would conform to a standard methodology and fulfills the key elements required to process a claim encounter. For example, the following led towards standardization: patient identifier, payer identifier, provider identifier, coding in the form of CPT, HICPICS(?), ICD-9 and revenue, authorization numbers and referrals for service.
A claim attachment provides the supplemental information to support a health encounter, usually formatted as a narrative summary or statement of results. The narrative nature of information requested in health care claim attachments presents a challenge to the development of a standard.
The content includes dictation of events, review of systems, treatment planning, patient responses to therapeutic treatment and results of diagnostic tests.
What types of claim attachments do we currently request from providers? The types of claim attachments that we currently request from providers include encounter and office notes, operative reports, emergency room treatment records, radiology reports, cardiology analysis, laboratory results, both diagnostic and pathological and referral and authorization forms.
When required to process a claim, the request for information is made via a system-generated letter. The letter is generated on paper and mailed to the provider. A series of three requests for information is typically made over a nine week period of time.
We do experience a 50 to 60 percent response rate within the first two weeks of their request for information. The information that we receive from the provider is usually received on paper, either via the mail or fax transmission.
Other types of documentation that we request are treatment plans for case management, physical therapy, speech therapy, occupational therapy, chiropractic care and medical records. Providers are advised of the requirement to submit these plans as identified above. When requested and not received, the request for information is made either via a telephone call to the provider's office or via a system-generated letter.
Again, the paper from the provider's office is received via the mail on paper or via a fax machine. Often, the nurses do take telephone calls for treatment records from provider's office. Needless to say, very time-consuming and labor-intensive.
What purposes do these attachments and other documentations serve? The intent for requesting supplemental claims information serves to validate the performance of services billed, review patterns of care, validate patterns of abuse, identify inappropriate procedural coding and assess patient response and progress to medical treatment options.
For example, when aberrant patterns of billing of care are identified, records are requested in order to review and advise the provider. A review typically generates provider education on procedural coding or counseling regarding concerns of quality of care issues by a plan medical director.
What aspects of these processes would be aided by standardization and electronic interchange for information. The processes associated with health care claims attachments that would be aided by standardization and electronic interchange are the initial request of information from the provider and transmission back and numeric laboratory data and referral and authorization forms.
What aspects of these processes are not conducive to standardization. The aspect of these processes that we see as not conducive to standardization are those that are associated with information that is stored in a narrative or non-standard format.
The documentation received following a request is often handwritten treatment and office notes, operative summaries and laboratory results. The relationship between a health care claims attachment and the medical record is that a health care claim attachment is a subset of the patient's medical record.
We do request entire medical records at times. The request typically is generated for the evaluation of an organ transplant, the identification of patterns of care, frequent emergency room visits, weekly office visits to providers, to review patient care outcomes, readmission to facilities and outcomes following treatment, the identification of quality of care issues, surgical complications, case management, to review the progress in chronic and acute cases, HEDIS medical records auditing and retrospective billing analysis.
If we were to suggest what organizations would have the responsibility for approving these claim attachments, the National Uniform Billing Committee and UBC form to develop and maintain a uniform data set and format for the transmission, charge and claim information to payers and third party from institutional providers, AAHP believes that the NUBC should have the responsibility for approving those claim attachments.
The National Uniform Claim Committee, the NUCC, organized to promote the development of a uniform electronic claim for use by the non-institutional health care community to transmit related claim and encounter information to and from third party providers, AAHP believes that the NUCC should have responsibility for approving those claim attachments.
If there was a governing body over data content of the standard attachment, AAHP would recommend that an incremental approach be adopted to adding the existing data content of the claims attachment. There are already data standards in existence for replying to certain routine requests; for example, the HL7 standards on test results.
AAHP believes that efforts should be made to build from these established standards. Our greatest concern relative to standardization is with claims attachment requests. All too often, patient data -- the request for patient data results in great inefficiencies or breaches of patient confidentiality.
For example, it is not uncommon for a provider to attach a member's entire medical record when only a small component of that information is required for the transaction. We believe that protocols should be developed for addressing these issues and problems of this magnitude first.
This concludes my statement. And thank you again for this opportunity.
DR. COHN: Thank you.
We probably ought to take this opportunity to introduce a couple of the other members of the committee that have arrived during the proceedings.
One is Dr. McDonald and then Dr. Starfield, the chair of the committee and Dr. Lumpkin, the chair of the subcommittee.
DR. STARFIELD: All transportations were delayed.
DR. COHN: Are there any questions from the committee?
DR. MC DONALD: This is a hard one to ask because attachments are so protean. The assumption sometimes in some people's minds is an attachment is necessarily an available piece of a chart, when in a lot of cases an attachment is a question completely out of the sky, I mean, from the perspective of some providers that has not yet been said anywhere else in the world and it is just sent off to the provider.
Are you thinking that it would be appropriate to catalogue and standardize those questions that aren't necessarily a common piece of the chart? I mean, we are talking about it would be -- I don't get all the letters right -- but the end committee's responsibility to deal with that and would they take that on or would the payers submit to such a constraint in question asking abilities?
MS. KORMANEC: As far as requests for information and standard questions?
DR. MC DONALD: Yes.
MS. KORMANEC: I guess I am trying to think of an example of --
DR. MC DONALD: Well, I mean, the attachments, I think, are falsely conceptualized as being this real contained, constrained little list of things and it is everything else, literally everything else. I mean, I guess some of you guys have been there.
So, we are really not -- these sort of form-like attachments, like for ambulance, but the questions on those varies across a spectrum that no one has quite got their hands on yet. I mean, the bigger companies, yes, but there is a huge spectrum of what people can put in there and it changes sort of freely. You know, this is America. So, you get a lot of changes.
Then there are those just questions that may read like just plain old questions, like why did you do this; "this" being, you know, why did you do that procedure and asking for a free-form answer. So, they are not necessarily -- some of these things require new constructed answers no matter what.
You know, you mentioned just sending the whole chart and that may be somewhat of a defense against having to interpret questions on the part of the providers. It is well chewed, I know. Others give you the whole thing and you figure it out.
DR. MUSCALUS: One comment I would have there is I think it is reasonable to try to catalogue in a question form the types of information that would be needed. However, I speak, again, as a contractor with the Federal Government in Medicare Part B. Because of our obligation to pay only for medically necessary and covered services, we certainly would not want to be constrained by a specific list.
There may be information that we would need that is not included based upon that list of questions. So, at the very least, there should be a process to allow additional questions or other items to be added to that list. So, I think the concept is good, but we need to make sure that before we make payment, we have reviewed any and all information that may be necessary to make sure that that service is covered and medically necessary.
DR. MC DONALD: I would just like to make sure the committee -- it registers with the committee that there is
-- even amongst the federal, most cooperative and sort of controlled part, there is this wish not to be constrained, which is kind of contra to the standardization process.
DR. COHN: Jeff.
MR. BLAIR: Let me try to piggyback on Dr. McDonald's question. In the process of trying to develop the standards, if it became clear that in many cases payers were viewing this as just simply a way of automating the request for claims attachment information that exists today and that they would just be able to get it a little faster and maybe a little more information.
So, I am going to ask you, I think, a little bit of a difficult question is: Have any of you envisioned how you could go beyond just simply using electronic claims attachments to ask more information and get it faster? Do you envision that this technology will enable you to alter your behavior in any way, where you are either more precise in your questions or your behavior would allow you to achieve efficiencies in any other way than getting the information faster?
MR. ANDERSON: I think it is possible to become more focused with the questions that are asked. It requires up front or to categorize the questions, but I think it is possible and desirable to focus it and reduce the amount of information that is going back and forth where it isn't needed.
DR. MUSCALUS: I think that our behavior may be altered with such a process from the standpoint of looking to automate more, to verify medical necessity more. It would allow us to develop certain medical policies that currently are somewhat difficult to develop because we would be able to obtain specific information that is unique to a particular case or claim that currently we have to develop using faxes or phone calls.
So, I certainly think that there would be a change in our behavior. I think from the standpoint of the Medicare program, it would help us to better protect the trust fund with regard to making sure that we are paying for those medically necessary services.
MS. KORMANEC: I think initially what we looked at as far as health care claims attachments is that information that we are requesting. I think as an industry, we could see it change our behavior. I would find it very hard to get a standard set of questions for the information because it would be unique from one carrier to another, although we would strive to make it more efficient on both the provider and the health care payer side to have consistencies amongst the information that we do request.
But I would find it very difficult to get to that final set of information that for each encounter would satisfy requirements across the board.
DR. STARFIELD: I have a quick question and then an observation for your comment.
Is it true that you basically review all claims and claims attachments rather than a sample of them, a random sample or do you review samples? In other words, when you -- do you look through your claims and say this one needs more information, we will request more information or do you do it randomly, say, every 10th claims, we request claims -- how do you decide what to request more information on?
MS. KORMANEC: Well, we review all of our claims transactions from a procedural coding perspective and there are things from that perspective that are kicked out and identified that you would want for further review and information. People are probably familiar with the Modifier 25 and if a provider attaches that to an evaluation of a management service, that means that I have done above and beyond what would be considered, you know, part of this service that I performed in the office.
Some payers will say fine. Other payers really want to say is there -- have they really gone above and beyond. So, there are instances and when we identify patterns in our different markets, you know, we will put a provider on, say, a full review of his medical records for claims submissions. So, some is triggered by pattern. Other is triggered by -- that we request that information.
DR. MUSCALUS: Similarly, what we don't do, at least in my understanding, is decide, well, today we will look at every 50th claim or every 100th claim and make sure all the blocks are filled in correctly. Our system is set up in an automated fashion that as long as a claim is submitted with all of the information necessary, we will process and pay that claim.
Now, certain types of services may suspend for review and that may be based upon program memorandums that are issued by the Health Care Financing Administration. It may be based upon a medical policy or we may have specific providers that we have identified and suspend their claims for specific review. But it is a completely automated system and it is not based upon setting up a certain pattern to review on an ad hoc basis.
DR. STARFIELD: Okay. The observation now.
I mean, it strikes me that this is absolutely the most extensive, implicit quality review that exists in this country. We are doing this with a system whereby records are linked by some patient identifier without procedures for protection of confidentiality or security procedures.
Is that a fair observation?
DR. MC DONALD: A comment. I don't think this is quality review at all. There actually is a whole other branch of HCFA that deals with that and this is strictly trying to make sure that --
DR. STARFIELD: But it has been described by all the testifiers as using -- as quality. Now, it may not be quality as we think we think of quality, but it is justification for procedures.
DR. COHN: I guess one would observe that the focus of this particular work is on payment, about whether payment should be --
DR. STARFIELD: I understand that, but it is still justification and that, in a sense, is quality.
DR. COHN: Dr. Muscalus, it looks like you want to make a comment.
DR. MUSCALUS: Well, I think that it is quality from the standpoint of was the service medically necessary. One could imply that if the service was performed and paid for, that really wasn't necessary, then that would implications for less quality.
DR. STARFIELD: But the point is is that it is being done with linked records. You link the claims and you link the medical record, which you frequently get and there are no standards for maintaining confidentiality and security, by and large.
DR. MUSCALUS: Well, again, I am not greatly knowledgeable with regard to our contract for the Medicare program, but I know that we do have certain requirements with regard to patient confidentiality that we have to adhere to.
DR. STARFIELD: Like what?
DR. MUSCALUS: Well, for example, we could not release information -- we may get requests from health plans with regard to certain policies or procedures that we have in place and we have to be very careful about the types of information that we release. Some information is releasable under Freedom of Information, but we have a process that we must go through before we can release that type of information.
I understand your concerns and I do appreciate them.
DR. COHN: Did you have a comment on this?
MR. ANDERSON: I was going to agree that the information that is received in paying the claim is not publicly available and must be protected by United HealthCare and other payers and it is not available to everyone within United HealthCare, only those people who are involved in adjudicating that particular claim and have a need to see it would be authorized to have access to that data.
DR. COHN: Great.
Kathleen.
MS. FYFFE: I have a couple of questions for all the panel members individually.
First of all, in general, can you say what percentage of claims you would request additional information for? And, secondly, could you describe what your internal management policies and procedures would be to maintain confidentiality of health information and also make comment about what you need to do in terms of adhering to state laws that relate to privacy and confidentiality?
MS. KORMANEC: I would say that we request additional information in 5 to 10 percent of our claims encounters. As far as maintaining the confidentiality of the information and our member health plan employees, who would handle that type of information are required to sign, you know, statements of confidentiality. Information that is stored on our system, again, is maintained by secure system access to information.
Your average health plan employee would not have access to clinical information for a member. They may just be able to access initial eligibility information. Are they covered? Are they not? So, we have different levels of security that would maintain that medical information that a health care representative would enter, you know, onto the system or transact as far as the claim was concerned.
DR. LUMPKIN: Can I just interject? You said 10 percent, but is that -- what about the four procedures in surgery? What percentage are those? Because that is a different class.
MS. KORMANEC: You mean operative reports?
DR. LUMPKIN: Operative, in-office procedures.
MS. KORMANEC: The 5 to 10 percent would be claims across the board. The request for operative procedures, I would probably put operative and office notes at about the 3 to 5 percent.
DR. MUSCALUS: It is my understanding that approximately 20 percent of all our claims require some additional information. I can get a specific number for the committee if you would like that.
Also, excuse me, with regard to any internal policies or procedures that deal with confidentiality, I could also get that specific information for this committee. The information would deal with requirements based upon our contract and also based upon our own corporation, our integrity and compliance program.
MR. ANDERSON: I don't have a specific figure on the percentage of attachments. The best number that I could sort of get as people's gut reaction was something less than 10 percent. And in terms of the security issues, we have an employee code of conduct that everyone acknowledges, which includes privacy.
Those people who work in the Medicare program actually are required to sign the Privacy Act just as government employees are. Our systems maintains and secure data using an I.D. and password arrangement and the facilities in which we work are secured and access is not given to the general public.
DR. LUMPKIN: I apologize for not being here for the presentations. Unfortunately, United Airlines out of Chicago was delayed by some stuff we call fog.
I will have to make a little bit of admission. I began this hearing Saturday night. I was at a fundraiser for a hospital in our community and there were a number of providers there. So, I used that as an opportunity to talk to a number of them.
Their perceptions are quite different of this particular process. Basically, the way that it was portrayed to me is that most requests for claims attachments only follow a refusal to pay. And then after a claim has been denied, what they perceive is then they have to -- well, why was it denied? Do I need to send additional information or is there some request associated with a claim denial.
So, the perception is is not only is there a frustration because of not knowing what the companies want, but, in fact, that delay in payment incurs cost to the provider and certainly at that time the payment accrues interest to the person who has refused to pay and doesn't accrue interest to the person who has provided a service.
To that extent, I know a number of providers I talked to say that for all their operative procedures, they send a copy of the operative note as a matter of course to every single claim that they send.
So that as we talk about the process, it seems to me that we are not just talking about automating the process of sending in the claims, but also have to talk about the interactive process, which is a major loss leader for many providers because to the extent that we have a complicated process, the time between submission of the first bill and actual payment gets dragged out substantially along.
So, to the extent that we are charged with simplification and streamlining a system, I was wondering if you had any insight on how not just -- I mean, if we conceive of the claims attachment as just being -- automating this record, that is one piece, but I would argue that our charge may be to automate the process of adjudicating the claim requiring additional medical information that generally is not covered on the standard billing forms that we have already talked about.
Any response?
MR. ANDERSON: I would agree that we would encourage the committee to look beyond just those attachments that might come in with the claim up front and to include the process of requesting additional information electronically and then receiving a response to that.
DR. MUSCALUS: I agree and I understand that providers would like to be paid in a timely fashion. It is my understanding that within our system we do not generate a denial first and then request information. We may identify a claim; for example, a surgical case with a 22 modifier, indicating there were unusual or difficult circumstances.
Yes, if there is no operative note attached or some other information to indicate what was unique about that procedure, why was it difficult, what was it that caused you to spend more time, we will suspend the claim, send a letter back to the provider and ask them to send us additional information.
If a claim arrives the very first time with an operative note or other information attached, there is a process that we follow to review that operative note, along with that claim to make a payment decision. So, when an attachment arrives, it is reviewed.
MS. KORMANEC: We are sympathetic to the need and, you know, that the provider feels that there are continually requested for additional information. As far as operative reports are concerned, they would standardly only be required when the provider bills an unlisted code of which we would need to derive a reimbursement or, again, for the Modifier 22 situation, where the provider is asking for additional reimbursement above and beyond the established reimbursement.
I can comment, too, that a lot of our denials to providers up front are related to their procedural coding errors, mostly in that the diagnosis does not support the performance of the procedure. We have implemented new standards within our system and have a high percentage of denial for codes not being supported by appropriate diagnosis, of which those generate the paper requests on the denial or the review of supplemental information, which usually turns into that things were not coded appropriately.
DR. LUMPKIN: Thank you for those comments. It seems that what we are describing as a situation whereby a claim is submitted through some adjudication process, requirement for additional communication, that actually doesn't cover the full extent of the use of a claims attachment. We may want to also look at the fact that prior to surgery there is a process of a prior approval of the claim, which is called prior authorization, which requires submittal of medical information. In fact, very likely much of the same medical risk history, history and physical and so forth, which currently does not occur electronically, but might in the same sense.
So, to the extent that we can look at automating this process, we may want to look at additional automations that will allow us to facilitate the process of actually providing care.
DR. MUSCALUS: One comment is with regard to applying something in the uniform fashion, right now the Medicare program does not define nor provide for any type of a preauthorization or precertification process. The committee just needs to take that into account.
DR. LUMPKIN: But your vendors do? I mean, the Medicare managed care organizations for which you are on a capitated rate, would do the preauthorization.
DR. MUSCALUS: Again, I am speaking as a Part B carrier in the fee-for-service environment, where those preauthorizations do not exist. I agree that with the risk HMOs, those processes are in place.
MS. KORMANEC: Right. In our Medicare risk, you know, there would be the prior authorization, you know, of procedures and the requirement for data.
DR. STARFIELD: Is your work load falling as you
-- oh, sorry.
DR. COHN: Jeff is next.
DR. STARFIELD: I just wanted to clarify his point, though, about the prepayment, the capitation. Is your work load falling as capitation is increasing?
DR. MUSCALUS: No.
MS. KORMANEC: Capitation provides your need to focus in different areas as well. We have issues surrounding capitation. Most of our capitation is in the primary care arena when your investigation of claims and services often stems over to your specialty care arena.
DR. COHN: Okay. Any other comments before we -- okay, Jeff.
MR. BLAIR: I always seem to think of questions that may not have real answers yet. But I will give it a try and see what you can do to answer this.
I think it is probable that it is going to take us some time before we are able to automate all aspects of providing electronic claims attachments. And you have observed that to some degree. You wound up realizing that we can't codify the narrative portions of the patient record.
Is there certain information, which would be low hanging fruit, where if we could prioritize the information that gets standardized -- since we can't do it all -- what information would either save you a lot of analytical time or save the provider a lot of resubmission of information? Is there anything that should be focused on as the highest priority in terms of standardization of health claim attachments?
DR. MUSCALUS: I will attempt to give as real of an answer as I can.
I think it would be good to focus on areas that deal with medical necessity; in other words, allow the ability to include diagnosis information. In general, that is already readily available, but also, I think, to tie into the billing system -- I am thinking in terms now of the CPT codes and the modifiers, where if you submitted a claim with a 22 modifier, you will automatically need additional information.
If it is tied into that process where we already know that when you bill this CPT code and attach this modifier, you will need to submit additional information. I think that will at least form a framework within which to operate.
DR. COHN: Peter or Rosemarie, comment?
MR. ANDERSON: Only to say that I agree with the concept of picking that low hanging fruit first. However, I would defer to the group I think you are going to hear this afternoon about -- I think some work has been done to look at the frequency.
DR. COHN: Okay. Rosemarie.
MS. KORMANEC: Yes. Again, I think that there are certain instances that we could identify, you know, out front, as I described in some previous answers with 22 modifier situations or 25 unlisted code situations. As far as the attachment information is concerned, I mean, it is all important to us. So, it would be very hard to say go with this first.
DR. MC DONALD: I have two questions.
One of them is -- I mean, this -- I see patients sometimes, so, I am sort of a provider, but I don't actually have to deal with attachments personally. Someone else does it.
And I guess the other thing is it may be common knowledge that physicians are paranoid about attachments. I think you understated it, John. They believe -- and I know this isn't true because I have worked with a lot of payers on this -- they believe that the only reason to put out attachments is delay payment and make the interest stiffer. This is pretty intensely felt, just as a background.
In terms of connecting to the patients' identifiers, you don't have to worry about that being too tight. My brother, who is in private practice, sends the things certified mail now, so he only has to send it once, because on the succeeding, they don't -- they don't get them connected always perfectly. But the questions where there is this precertification and postcertification -- and this was a suggestion made at one of the pilot meetings in Chicago by a provider from New York -- these are really two different kinds of requirements and when they get tangled together, it really causes a little confusion.
The precertification is mostly medical necessity, which isn't really too tough. The data needs aren't really unmet. And then postcertification is really -- that is the bad buggers; you know, it is one of these sort of things. As I understood a lot of that, you really can't do that without being able to go in and kind of roust about in the clinic and in the office itself anyway.
That is, you have to kind of boil it down to what are they really doing exactly and it may not be very amenable to electronic stuff at all and it is sort of -- and if we separated those out, would we have an easier time automating the precertification and just leave the postcertification? Some people, say pay me and then come back and put me in jail if I have mis-billed. It would be easier in the long run.
Well, the sense really -- I didn't mean -- but the issue really is distinguish those two issues, as if you really are trying to stop someone from defrauding the government, then those things you can pick up in different mechanisms and it does require fairly intense data review, much of which has to be looked at in the original hard copy.
DR. COHN: Comments from the panel?
MS. KORMANEC: I don't think our primary goal on claims attachments is to get the provider and put them in jail for inappropriate coding. We are basically looking at providing good quality care, you know, to our members.
DR. MC DONALD: I think I misdirected the question. I really don't -- no one is believing that. The question is there really are two different dimensions to the attachments, two different needs and uses. One of them is to check for fraud and abuse. And the other one is simple medical necessity. I think it is referred to precertification and postcertification if I know the jargon correctly.
If one said we would try to automate all the precertification and let the payments flow on that basis and leave the postcertification as a later project to automate because it may need coming in and taking pictures and, you know, interviewing people and that sort of stuff anyway. So, it isn't as amenable to automation.
DR. STARFIELD: Could you explain what you mean by pre and postcertification? I am not sure what you mean.
DR. MC DONALD: That is just a jargon I picked up. But there is -- it is basically -- medical necessity can boil down to -- some could argue all you need is a diagnosis and a CPT-4 and you can automate the connections and that defines "medical necessity."
Now, you could stretch another couple yards or maybe a couple of hundred yards and it still wouldn't be tons of different kinds of data. But the postcertification opens up all questions on the universe and, in fact, as I understand it -- this is just picking up from the meetings I have been at -- is that it often requires a physical audit on site when people are worried about things, postcertifications. So, it isn't as amenable and if we would make a -- distinguish between these two kinds of data requirements, it might make life easier and then we can make progress faster.
DR. COHN: I myself would disagree with some of the conceptualization but I think the panelists have some comments.
DR. MUSCALUS: I think what you are referring to is prepay versus a post-pay review. And while I agree to an extent with your comments, I think it may work, as you described, in an environment where you are preauthorizing and precertifying services before they are performed. And if someone gets preauthorization, they perform services and get paid and there is a problem identifying, that may clearly become a fraudulent activity.
However, in our environment, the fee-for-service Medicare environment, there is no preauthorization. Therefore, we are required to look at our data and sometimes on a post-pay basis make a determination as to whether or not services were medically necessary.
One brief example, we had a certain provider who was rendering respiratory therapy services in nursing homes. Essentially what they were doing was going room to room, often just administering patients their own hand-held inhalers three to four times a day, seven days a week at about $30 each time; up to a million dollars a month, they were billing.
We on a post-pay basis requested medical records and based upon a review of those medical records, we were able to identify that some of these services either weren't medically necessary. They may not have even been performed or they weren't performed according to the right guidelines.
Now, often these cases do go on to fraud and investigations, but many times we will on a post-pay basis determine medical necessity and that can often be determined solely through the use of a review of the records and not necessarily by going out and doing an on-site review.
DR. COHN: Actually I have a question, too. Let me start out with mine since I have got the chair, so I get to ask an occasional question.
DR. LUMPKIN: It is a follow-up on --
DR. COHN: Is this a follow-up? Sure. I will defer to you as a follow-up.
DR. LUMPKIN: Thank you because I think I wanted to clarify one thing on post-review and having administered a program -- we in Illinois do WIC -- we are able to identify risk for fraud based upon electronic patterns. And it is those kinds of things, which may generate some additional review.
So, I think that, yes, there is a lot in postcertification that you can do. It was mentioned that certain codes automatically generate a response and that is based upon experiential experience and frequently the payer does not want to let the providers know what the pattern of review is because then the providers -- they are concerned that the providers may then model their billing, just as the IRS does not let people know what stimulates an audit because then people will modify their tax forms.
But there is one other piece to that and that is that in some of the providers -- actually, one provider I happen to know fairly well, who is an OB/GYN, it was her experience that if a patient gets a diagnostic laparoscopy and has endometritis and subsequently gets surgery for that and the plan does not cover as a benefit as a benefit, infertility, that they will review the record to see if there is a mention and somewhere -- not on that op note, but somewhere two or three or four months earlier in the medical record, a mention of infertility because the chief complaint may ultimately have been infertility, the ultimate procedure, and then they will deny the claim.
So, the process, which the payers go through is very complex and it is not always clear to the provider why they need so much information and it has to do with the very complex relationships of what is paid for and what isn't.
DR. COHN: Let me ask a very -- I am going to go back about two or three steps here because the reality is is that the claims attachment in one way or another is really at least in some ways a standard already. ANSI was a 257 X12 standard. Have any of you implemented electronic methodology for getting attachments at this point? Or is everything still fundamentally paper in all of your environments?
MR. ANDERSON: The closest that we have come is the step before that where we asked electronically for information. We did that. It is not a permanent installation that was done as a proof of concept to show that it could be done. We intend to be doing the 275 to take it back in, but that is not in place today.
DR. MC DONALD: I think it needs a clarification. The 275 is not a standard for any of the attachments. It is a shell.
DR. COHN: Right, but what I am saying is it could be implementable between partners.
DR. MC DONALD: Well, there is a semantic problem. I mean, if you look at ten ambulance forms, they all have different questions on them. That is not a standard. And when you stuff into a fax, into the 275, I don't think that is what you are meaning when you are saying what you are saying.
DR. COHN: Well, I was just curious to see if any of them were doing any aspect of this electronically at all. So, it was a really basic question.
MR. ANDERSON: A good analogy, I think, is that the 275 is almost an envelope. It is not the contents. It is a standard way to move the data back and forth, but each of the pieces of data or the attachments that might be within it also needs to be standardized.
DR. MUSCALUS: But right now, we do not have a formal mechanism -- providers can if they bill electronically include certain information on that electronic claim that will allow us to facilitate payment without having to request records.
Similarly, if they submit an electronic claim, they can at the same time send us a fax so that we will be able to know that additional information is in route and get the two of them together to process the claim.
MS. KORMANEC: We have not implemented any method to get that information.
DR. COHN: Okay. Peter.
MR. ANDERSON: Although we have not done the X12 275 transaction, there are other examples where what you might call an attachment are coming electronically today. An easy one would be in the durable medical equipment. HCFA's regional contractors uses a format, the national standard format. Part of that is certificates of medical necessity.
They can come in and do come in electronically today.
DR. MC DONALD: Just one other clarification because the claims themselves contain formal attachments data now. So, that there are actually fields for BON and hemoglobin and epidose(?) and a variety of additives for amino acids that are very specific. They are very specific checking algorithms, which are really conceptually in a class that we call attachments, but they are fixed little fields in the claim. It is a messy world.
DR. COHN: Kathleen.
DR. FRAWLEY: I would just like to ask the panelists, we have been talking primarily about physician billing. I was just wondering your experience in terms of hospital billing, in terms of claims attachments. Is the types of inquiries you make similar to what you would be asking a physician to provide or are there significant differences?
MS. KORMANEC: In the arena of facilities, we request a lot of ER treatment records, which would be from a facility. Requests for information, as far as like inpatient utilization, is generated for bills of certain dollar amounts, of which would require a full audit of the patient chart.
There seem to be less requests on the inpatient facility side than there is provider or office-based.
DR. MUSCALUS: My experience comes as the acting medical director for Veritus(?) Medicare Services, which is the Part A fiscal intermediary. Most of the reviews that I have been personally involved with have dealt with nursing facilities, either on a prepay or a postpayment basis to determine whether certain services, such as occupational therapy, physical therapy, speech therapy, were, in fact, medically necessary, whether they were services that required administration in a nursing facility.
So, those are the areas where I have been involved and the other is a review primarily of the medical record.
MR. ANDERSON: In preparing for this, the answers I got from our Part A and Part B Medicare would lead me to believe, at least, that there is less frequency on the Part A, the institutional side, than on the professional.
DR. COHN: Jeff.
MR. BLAIR: Another question that may be hard to answer.
Does anybody have any feeling for what percentage of the time -- the question you asked back to the providers are open-ended, like why did you do this procedure or why did it take so long, whys and hows, as opposed to asking for a specific report or code or document.
Let me tell you why I am trying to get at this question is because -- and I don't know what the answer is, but if there is a fairly large percentage of whys, then the providers are going to have a real hard time isolating the kind of information to send you back in any kind of a standard and they will probably wind up just automating the fact that they send you as much of the record as possible because they don't really know what you are looking for in terms of an answer. They will send you the whole progress note or operative or whatever it is.
Do any of you have a feeling for what percentage of the time you are asking questions like why instead of a specific piece of information?
DR. MUSCALUS: Again, as a Part B carrier, we tend not to ask the questions, why. I think, in fact, the answer to that question has implications for quality of care issues and while those questions may ultimately have to be asked, they may best be asked through the peer review organization or through the quality assurance committee of a managed care organization.
Rather, the question we ask is please -- in fact, it is not necessarily a question. Rather, it is a request to provide us with the documentation, which describes what it is that you did. That information is then reviewed in certain circumstances by a peer of that individual to determine whether or not what was actually documented reflects what was billed and whether or not it was covered, medically necessary and, therefore, payable.
MR. BLAIR: But it is still pretty much a general open-ended, provide us with the documentation that describes it.
DR. MUSCALUS: We are asking for a document. That is correct. We are not asking for an explanation. Rather, we want a document that describes what was done.
DR. COHN: Peter? Rosemarie?
MS. KORMANEC: Well, I think the request would be specific. Provide me the office notes for x date of service. We need to see the operative summary. We need to see the lab encounter data. I would not see us asking the why and the where.
DR. MC DONALD: As one who read about 500 attachments, including ones in large volume sent to our hospital, I don't know from whence or where, there were questions like why did you do two of these procedures in the last six months. I mean, there is not a record in the chart that you can yank and say this is why. And there was one I remember, which was kind of cool. It said why did you two things two days after the patient died.
Now, probably, you know, the dates were mis-coded but I mean, you might have been able to turn that into a code, but there is a number of them that are sort of open-ended and in within even the standard forms, the standard documents, which are these sort of like questionnaire forms for 10 or 12 different kinds of things. There is usually a question and it is fairly open-ended, which requires manual labor. It is the point, I think, Jeff was saying is the payers are worried about how can we stop this flood of paper and the providers, I think would like to be able to yank it out of their record systems, but if all the payers give them is open-ended questions, they are still going to need a manual person in between to review their chart and make up an answer.
DR. MUSCALUS: One issue, we also have to be careful that -- going back to the issue of infertility and endometritis, that if a claim comes in and either is denied or suspended and we go back and say, please, provide us with a copy of the -- either the operative note or the preoperative history, you may do that. It will be reviewed and still information isn't enough to make a payment decision.
We then go back and ask for more information. So, generally, our position is to go back to the provider the first time and say, please, provide us with the documentation to support this. It may be the operative note, but we give the provider the opportunity to provide us with as much information as they would like to support the claim that they submitted.
So, again, you just have to be careful. If you ask for a specific, you may get specific information, but it still may not be enough and then further delays in payment.
MR. ZUBELDIA: Kepa Zubeldia with ENVOY-NEIC.
I have a why question for the panel that kind of follows on Jeff's question on the low hanging fruit. It does seem maybe that 30 and 40 percent of the claims have a secondary payment involved from a secondary payer with the exception of maybe care secondary and maybe the crossover is between Medicare and Medicaid.
There is very few, almost none of the payers that accept secondary claims with the electronic attached explanation of benefits. The 837 and -- I mean, the 837 provides a very good way of doing this crossover of claims with secondary request for payment with an attached explanation of benefits.
Why is it that the payers are not accepting this electronic explanation of benefits and they insist on having a paper copy of the EOB?
DR. COHN: Would anyone like to respond to that?
DR. MUSCALUS: I am looking back at my technical support team and I am being told that we can, at least, at Xact, we can accept the electronic EOMBs.
MR. ANDERSON: The answer from a technical perspective, I would say that it is not -- it does not turn out to be a technical problem to do that. It turns out to be more of an office problem and people's familiarity and, you know, liking to have the piece of paper and being able to read it, unfortunately.
MS. KORMANEC: Being from a non-technical background, I don't think I am going to go there.
DR. LUMPKIN: Well, I think that question, if I understand it, begs another question, which is which is more economical for a provider, to just have somebody xerox a record or to -- because most of these are not kept. The vast majority are not automated -- and just send that in or to try to abstract it and respond to requests. I would suspect most providers would just rather have someone in their office xerox the paper.
DR. MUSCALUS: We have some providers who provide various services and they know that -- or they have learned to expedite payment. They automatically send information right along with the claim. So, providers are already doing that.
DR. COHN: Speaking of that sort of as the follow-up, how much of the information you get as claims attachments are things that look like -- I mean, that are typed basically as opposed to handwritten or forms that are filled out by hand or whatever? Is this -- what is the percentage? I mean, if it is typed, it can be computerized. Things that are written by hand typically is going to be sort of a tough one.
MS. KORMANEC: I think for the most part your office notes are handwritten, of which you have to decipher and can only read about 1 percent of the information typically. So, we would love typewritten, electronic methods of getting that information.
So, I would say office encounters are typically handwritten responses, your operative notes and your laboratory information is typically a typed submission.
DR. MUSCALUS: Again, most of the handwriting that we see comes from office notes, in the hospital, progress notes and nurses' notes. If in the course of a review we find it difficult to read an entry, which, as you know, is somewhat of a common finding, we can go to a provider and ask them -- we can ask a provider to send in a transcription typed of what was written.
Now, obviously, if what is written is two lines and we get a full page typed, that creates some concerns, but, nevertheless, they do have the opportunity to help clarify for us what is in the record.
DR. COHN: Jeff.
MR. BLAIR: I keep going after this very difficult question, so, I will try it from a different perspective.
Obviously, you maintain staffs to review claims and when you don't feel comfortable, you ask for additional information. It costs you money and time to maintain those staffs and if electronic claim attachments were available to you, assuming that you are hoping this would help you become more efficient and reduce those staffs, do you have any feeling for what types of questions, what types of reviews are the largest demands, which cost -- where you have to maintain the staff?
What I am trying to get to, again, is some other way of feeling where the low hanging fruit is. If you are spending an awful lot of money reviewing certain types of claims, maybe that is where the priority should be placed. Okay? Maybe the priority isn't a technical one or a numbers, but it is in terms of your costs of reviewing things. Where do you spend the most money on reviews of claims?
DR. COHN: Jeff, I need to have a little -- I have a question for you about this one because it seems to me that even if they get it automated, they still have to have a staff a review it. Is that correct?
MR. BLAIR: Sure.
DR. COHN: So, you are just saying that the issue is having to do with getting the paper themselves.
MR. BLAIR: I am still trying to find out if it is possible to know -- if we do have to prioritize and focus on what aspects we automate first, what will in your case reduce your costs the most.
DR. MUSCALUS: One of the things that I did was actually go over and sat with our individuals who look at these attachments to get a better feel for what it is that they do on a day-to-day basis. Some of the examples that I saw were claims with 52 modifiers, which reflect a reduced or an incomplete procedure; 62 modifiers for co-surgery; codes for multiple visits, whether an outpatient or an inpatient; 53 modifiers, 22 modifiers and also claims with NOC are not otherwise classified codes.
Other areas where we do reviews are in determinations of cosmetic or reconstructive surgery. Also, we have flagged providers where before we make payment, we review as much information as we can. Right now, we are heavily involved with certain ambulance providers and looking at those types of claims.
Also, if one of our analysts cannot make a determination based upon the information they are reviewing, they then would go to one of our medical director assistants to get some additional information, rather than -- my understand is rather than immediately go back to the provider who submitted the claim. In other words, there may not be enough information and they would go back to the entity, who submitted the claim, but they do have the opportunity to go a physician in order to get an interpretation.
DR. LUMPKIN: Let me, perhaps, refine the question because it may be a little bit unfair to ask you today to answer that question, but could you provide the committee with some statistics on what generates or what leads to requests for a claim attachment? You just listed a whole group of those. But would it be useful for us to know -- we heard a range between 3 and 20 percent or those which claims attachments are requested -- it would be useful if we could get some counts of what are the most frequent causes, other than 25, 52.
I think we heard a couple of those. I think that would be very helpful for us.
DR. COHN: Okay. Other comments?
MS. KORMANEC: I think the electronic receipt of the information would not decrease our costs associated in the areas that you need to have clinically-based qualified people review the information, as far as the electronic receipt would make it more efficient on our clerical supportive requirements.
MR. BLAIR: That is the answer that I was afraid you would give.
DR. COHN: Peter.
MR. ANDERSON: I was just going to say that I don't have any percentage, but I suspect that we ask a fair number of questions that are not asking for attachments in the sense of send me this form or that record, but more discrete pieces of information that could be machined when they were returned, so there would be some benefit from doing those.
DR. COHN: Mr. Moore and Clem, do either of you have a follow-on or is this a --
MR. MOORE: It is to do with -- what you do with the attachment when you get it. I am Bob Moore from HCFA. And you mentioned the certificate of medical necessity that comes in on DME. How much of those really are looked at by staff? Like a lot of attachments are sent in, is it the sentinel effect that we are creating, that we get a lot of paper but we don't look at all of it? It is there and available to be used in the adjudication process.
Do you know, to your knowledge, you know, do we look at every piece of paper we get on the claims attachment by a person? Do they analyze it to see if the services are really necessary or is it a matter of routine that we request it and we look at a portion of it?
MR. ANDERSON: As far as I know, if we go to the trouble of requesting an attachment, it is with the express purpose of having somebody review it when it does come back in.
DR. MUSCALUS: I also get the sense that we have to request it because we need it before we can make a payment determination. I do not get a sense that these attachments arrive and they just sit somewhere and no one ever looks at them. My sense is that we do look at that.
MR. MOORE: Some payers have indicated that when they go back to their claims review people, they find out that the attachments have never been reviewed, but they have paid the claim. But it serves as an alert to the provider that we might be looking more deeply into this particular service and, therefore -- and they will screen some thoroughly than another.
It does create a volume in the provider side.
DR. COHN: Interesting.
MS. KORMANEC: I think that we try to request attachments and supplemental information for, you know, only those times when we really need the information to process, of which we have the procedures in place that would set further review prior to payment.
MR. MOORE: I am thinking more of those, like the certificate of medical necessity and some of the other services --
MS. KORMANEC: We don't have that.
MR. MOORE: -- that the provider is put on alert, don't send me a claim unless you have this attached to it. If you do send it to me without this information, I won't be -- you know.
MR. ANDERSON: In fact, on the Medicare side, with which I am most familiar, those are actually a part of the automated process and they are checked automatically and they have to be there and they have to have the right values in order for payment to be made.
DR. MUSCALUS: Additionally, it may be that there are certain requirements in Medicare that says before a carrier can pay, this must exist and, therefore, while it may not be thoroughly reviewed, nevertheless, the provider is required to include that information when they send in the claim.
One other comment I want to make -- Dr. McDonald commented about perceptions and that we ask for this information because we want to delay payment. One of the things within the Medicare arena that helps, I think, the providers is the requirements that contractors have to pay claims in a timely fashion. And there are specific parameters that we have to process electronic and paper claims and we have specific days in which to accomplish that and if we don't meet those requirements, then that is an area for discussion at the time of our contract renewal.
So, having those guidelines there for us is incentive for us to make sure that we don't let things pile up and delay payments back to providers.
DR. LUMPKIN: Let me clarify that statement. Does that mean that when you suspend the payment, the clock is still ticking?
DR. MUSCALUS: I will have to go to my technical support people on that one.
Julie, do you know?
We think it may, yes.
DR. LUMPKIN: Okay. So, if you ask for a claims attachment, it doesn't count against your performance for Medicare?
DR. MUSCALUS: I don't believe --
MR. MOORE: I think that is a clean claim --
MS. KORMANEC: A clean claim.
MR. MOORE: -- that the time requirements -- so, the claim gets suspended and the additional documentation required, that doesn't count in the timeliness factor.
DR. MUSCALUS: We could get clarification, if you would like.
DR. COHN: I would rather be suspended than denied any day.
DR. MC DONALD: I just want to point that we have got a -- we don't have the usual information system design context, where both sides have a clear idea what the data elements and they want to move them back and forth. We actually have more of a, you know, theoretic game situation. I don't mean this in a negative sense. People want to, you know, cloak a little bit of what they are doing so that they get certain incentives, motivations back and forth.
You know, the whole notion of we are going to do
-- you know, if we could get to the discrete data elements and you could anticipate on the provider side needing them, you would store them and we could make this system a lot easier, but as long as it has to be sort of surprise and stealth -- I don't mean that in a negative way, but like in a military way, I don't know how we can ever help with automation.
So that I think we either have to get over the thing and use other methods to sneak up on them, you know, and catch them in bad stuff, but really clarify what it is we are doing so they can put it in their system and you can have your system and it can all be done automatically.
So, there is a sort of fundamental conundrum here, I think, that we have to at least say it is there and admit it.
DR. COHN: Thank you.
Any comments?
DR. STARFIELD: Have any of you looked at any of the systems used internationally like, you know, in those Western European countries that pay fee for service for things, you know? Do they go things like this?
DR. MUSCALUS: No, I haven't seen any.
DR. STARFIELD: Do you think it makes a difference if they don't do that? I mean, from everything I know about, you know, their health statistics are better than ours, but --
DR. MUSCALUS: I don't know.
DR. STARFIELD: And the costs are lower.
DR. COHN: Well, any comments there? I actually have one -- is there a comment to that? Okay. We are running a little late now. I actually have one final question and then we will adjourn for half an hour.
And that is just from your view because we will -- I think we will end this particular panel right now, as opposed to coming back after the break.
And that is as you have listened to us and the questions we have had around claims attachments -- I know that all three of you have thought a lot about this -- what advice would you give to the committee as it advances into this area, either words of caution, first things that we ought to be doing, suggestions?
I would ask for comments from all three of you.
MR. ANDERSON: I think the approach might be to not try and do everything at once, but try and take steps in the right direction, start with discrete attachments. In my mind, making an attachment too big is not the way to go. The more you can isolate it to one specific use, the better, one set of data, the better.
My suggestion, I think, would be to try and do that and over time add more types of attachments to what you have handled and maybe build on the experience you have early.
DR. MUSCALUS: I agree. I think because of the implications of this type of a system, it is important to go cautiously and to be as thorough as possible. I think, also, identifying those areas that are common and focusing there, such as has been suggested, would be beneficial. I think it is important to fully understand the differences that exist among payers.
We have a contractual arrangement that we have to abide by based upon the Medicare program defined by the Social Security Act. We can't bend a whole lot. So, what we do and how we do it will differ from a Medicare risk program, from a commercial HMO product. So, I think it is important to get a good understanding of what other payers, how they operate and information that they need.
My last suggestion is to continue to work with providers and payers, so that you can be aware of what their issues are. And I think if you go in that direction, it will make it easier for you.
MS. KORMANEC: Yes, we are concerned, you know, for the approach to it and we do want to take a cautious incremental approach. You know, we are concerned for -- once we get to a standard, you know, what are the processes to modify it and what is the impact of implementing those changes as we go forward.
We would look towards starting with the small things first and then moving towards the larger items later on.
DR. COHN: We will get copies of your statements and all that.
With that, I want to thank you. We will break for half an hour. We will start again at 11:00 --
DR. LUMPKIN: In about 20 minutes.
DR. COHN: 20 minutes.
[Brief recess.]
DR. LUMPKIN: Our next panel is "State and Public Health Perspectives." And I would ask the two panelists to introduce yourselves and also since this is going to -- this is being recorded for the transcript and also going out live over the Internet, please speak into the microphone. So, if you will position it in front of you, so everyone can hear you.
MR. DAVIS: My name is Bob Davis from the New York State Department of Health.
MR. LUNDBERG: Michael Lundberg, Virginia Health Information.
MR. DAVIS: I guess I will start.
My name is Bob Davis and I represent the New York State Department of Health where I am responsible for maintaining the Statewide Planning and Research Cooperative System, widely known as SPARCS. Clinical and financial data for all inpatient and ambulatory surgery discharges from New York regulated hospitals is collected from a system founded in the late 1970s, using NCVHS recommendations contained in the Uniform Hospital Discharge Data Set and continues to be used for rate setting, utilization studies and a wide variety of research projects.
I would like to thank the National Committee for inviting me here today and giving me an opportunity to share our experiences in New York State. The department is committed to data standards as the full impact of the administrative simplification provisions of HIPAA legislation evolve.
For us, attachment information is any information needed beyond what is required by any payer to adjudicate a claim. An administrative database, such as SPARCS, will always need certain data items not necessary for payment of a claim but no less important. To meet our needs without placing an undue burden on providers, our requirements need a home.
Deciding whether that home is in the claim transaction or the attachment transaction, we feel, should be dictated by logic. To illustrate my point, I would like to state two assumptions followed by two examples of how two of our additional data needs could be met and accommodated in the existing transaction sets.
Ground Rule No. 1: All standards development organizations, SDOs, need to accept the fact that administrative simplification can only be achieved when duplication and data redundancies are eliminated at the source, which are provider information systems. This requires SDOs to take a global view of information needs for health data.
Ground Rule No. 2: Every payer needs to recognize the fact that other data collectors, i.e., other payers and other state agencies, like SPARCS, have different requirements and they should not reject claims because too much information is submitted. Follow the axiom; ignore, don't reject.
Like, for instance, No. 1, in New York State, two additional discharge status codes are needed to ensure appropriate provider reimbursement has occurred and to enrich research. The two additional codes are for transfers of babies with low birth weight and transfers of trauma patients to regional hospital centers.
Currently, the National Uniform Billing Committee maintains an external code list and argues that these codes are not needed to adjudicate a claim. I would agree to that, but that does not lessen the importance of this information.
The question remaining then is: Is it logical for providers to maintain two discharge status coding schemes to support these additional and useful codes. I think the answer is definitely not. And the logical place for these codes is with the already-defined discharge status codes.
For instance, No. 2, as recommended by the National Committee of Vital Statistics core elements in the UHDDS before, New York State and California collect a diagnosis indicator to determine the presence or absence of a diagnosis before admission to the hospital. While the 837 transaction set provides for collection of other diagnosis codes, there is no room for these associated indicators.
A slight change in the implementation guide or minimal data maintenance to one data segment will allow for a logical and manageable means for collecting this information. Does it make sense for providers to duplicate all the diagnosis information just to avoid collecting this one piece of information for each diagnosis on the claim transaction?
Again, I say definitely not. There are, obviously, other data elements needed for the public health research systems that are not in the existing claim transaction, that do need a home of their own. The New York State Department of Health is hoping the attachment transaction set is robust enough to accommodate these additional needs.
The New York State Medicaid Program is also under the auspices of the Department of Health. In addition to these issues I mentioned related to SPARCS, the state Medicaid Program also has additional data needs related to the claim or ability to comply with state and federal requirements that are presently legislated or regulated.
As a participant in the process, I know one essential ingredient to make this feasible is an electronic staple between the claim and attachment transaction sets. With the ability to link various transaction sets, we feel the real issue of standardizing data content for all justifiable uses can become the main focus of SEOs in order to achieve administrative simplification at the source provider information systems.
I was fortunate to be invited to a meeting sponsored by the National Center for Health Statistics and the Centers for Disease Control on January 23rd, that brought together several organizations that shared a common need for additional data beyond the claim transaction, and that includes NAHDO.
It was very encouraging that there was a unanimous consensus at that meeting to move forward to sponsor a workshop that would broaden the base of parties interested in bringing standardization to these additional data needs.
I want to thank the NCVHS on behalf of the New York State Department of Health for giving us the opportunity to share our commitment to standards in a way to make better decisions on relevant health care issues of today and on into the future.
It is obvious from the questions from presenters I received, along with my invitation to speak at this hearing, that must thought has already preceded these hearings and there is a strong desire to solicit industry input to achieve consensus solutions to complex but important data issues confronting us today.
This effort will serve to better prepare the country for tomorrow's health crises in an affordable way.
Thank you, again, for your leadership and the opportunity to express our views.
MR. LUNDBERG: My name is Michael Lundberg. I am the executive director of Virginia Health Information. Virginia Health Information is a private, not-for-profit organization that contracts with the Commonwealth of Virginia to collect, analyze and disseminate information on hospital discharges, outpatient care provided by physicians, outpatient care provided by laboratories and others.
We also collect financial and performance data for hospitals and nursing homes. Our board of directors is represented by business, consumers, hospitals, health plans, payers, as well as state interests, nursing homes and physicians.
Virginia Health Information operates as a public/private partnership to create health care information, to allow consumers and employers to make better and more informed health care decisions.
I am also here representing the National Association of Health Data Organizations. NAHDO members include individual, group and private organizations, running the gamut from students, professors, state data organizations, those with modest incomes, to multimillion budgeted organizations.
We have a common thread throughout our organizations and that is to help develop and disseminate actionable health care information. NAHDO members support standardization. The UHDDS, the UB-92 and the UB-92 and their predecessors have led to laws and treatment approaches that touch our lives everyday. These include seat belt laws, child safety seats, studies on long term management of diabetes, cancer, stroke and the gamut of health care diseases that affect people today.
But existing standards in data collected is not enough. Information is gleaned that is used by consumers, employers and payers to compare health care providers. It is also designed to help improve the care that providers provide by feedback.
In response to Question No. 12, do you currently require providers to submit information that is not part of the standard claim, many states, including Virginia, collect additional information from providers in order to allow for the adjustment to account for the differences in the levels of severity of illness that patients experience and bring to a health care provider when they present.
Virginia collects information on patient race and ethnicity, patient social security number, infant birth weight, external cause of injury codes, also known as e-codes, the uniform physician ID number and Medicare provider numbers. Through collection of that information and the design of that, we always balanced the burden of additional collection with the value to the ultimate users of the information from them.
Many other states have other information that they collect for similar purposes. California collects up to 24 secondary diagnoses, information on conditions present at admission to allow distinction from complications arising during the hospital stay and others. In Massachusetts, they also include socioeconomic indicators, information on patient history, patient reported health status and others.
Now, many other states -- and there are 40 states that have mandates to collect health care information, primarily hospital discharge and outpatient -- many of those states want to collect additional information but they are looking for guidance and they are looking for standards in this.
Now, regarding Question No. 13, how do we intend to continue to collect information from providers that it is not part of the X12N 837 health care claims, there are some options. We see an urgent need for you to consider, as has been suggested to us by the Health Care Financing Administration, for claims attachments to be more broadly defined than is currently in the practice in such a way to allow for the collection of information necessary for health outcomes evaluation and health services research where the data is not necessarily for claims adjudication.
Now, it might make some sense, as Mr. Davis mentioned, to incorporate what you can in the existing claims form. It is also our understanding that much of those decisions have already been made and that most of the options now would exist within the health claims attachments. So, I can argue with the point of view that it would make some sense to add additional diagnoses to the claims form, but if it is a done deal, where there is not much opportunity for input, we think the claims attachment is one area to consider.
We have also sought confirmation from the Secretary of Health and Human Services that organizations collecting information from providers that are not conducting transactions, such as claims payment, enrollment and related correspondence, two-way correspondence, that they are not prohibited from doing so by the administrative simplification.
That remains a question that we are seeking clarification on that. Having said that, I will say that we would much prefer that the X12N 837 and the claims attachment be coupled to meet states' needs. It makes a lot of sense. It would avoid duplication. We could have standards across for definitions from the transaction.
So, toward that end, I would like to answer Question No. 14. Will states come together, in other words, under the aegis of the NAHDO, to design a claims attachment that will serve their information needs in a standard way? And the answer to that is "yes." The NAHDO has already assigned an Administrative Simplification Committee to identify the issues, to poll states on the different kinds of information they are collecting and to move towards consensus.
Now, from the comments I heard this morning, it was fairly clear that under some cases, providers submit additional information based on their own feelings and what they think will get paid. Payers in some cases request additional information in formats according to what they think will work and help them do what they want.
By the same token, we shouldn't think that states instantly would come to agreement on exactly what they would want in addition. There would be quite a bit of hand-wringing. There would be a lot of discussion. The only thing states would not want to lose is what they are collecting. I think we could all realize that. But it is an effort that NAHDO is up to. They represent the vast majority of those states and they would work together to try to do this in a fashion that would predate the actual claims attachments deadlines that are required.
Now, the final question is: How can this subcommittee help in this effort? I think there are a number of things that you can do. One is that if, in fact, claims attachments are chosen, that you ensure the claims attachments are defined so as to allow for the collection of information to support health care outcomes analysis and related information not necessarily required for claims adjudication. This is critical if this is the approach you wish to take.
Secondly, to ensure that once approved, that providers do collect and payers pass this information on. This should be reinforced by a consideration of a reasonable and rudimentary series of edits.
Next, in the schedule for implementation, we must recognize the absolute necessity to allow states current collection efforts to continue unabated until phase-in of all the claims attachments and related information is completed. We need to avoid gaps in the collection of this information.
Again, there were 40 states with mandates to collect health care data. They are experienced. They are knowledgeable and greatly appreciate your interest and continued consideration of their needs.
NAHDO does and will continue to represent these groups and constructively work with you and the Department of Health and Human Services. We appreciate this opportunity and the chance to answer your questions today and in the future.
On behalf of NAHDO, the board of directors and the Administrative Simplification Committee, I thank you.
DR. LUMPKIN: Thank you.
Questions?
MR. MOORE: The collection of this data costs something and most of the data flows through the payers from the provider unless, you know, some -- in many instances. But you had said that as we -- that you would like to see the committee support the collection of information that is not used in the claims adjudication process by a payer. How would we spread the cost of that if we are going to edit the data at the source and get it back to the provider or where the data is coming from?
Would you see that being the burden of the payers? Or how would you see the system working so that the burden doesn't fall to one of the participants in the system but would be borne by all?
MR. LUNDBERG: Well, I think there has to be give and take all the way down the line. I think the first thing to make clear is that these data generally are already collected at the hospital level. The information is already there. It is just a question of the extracts.
Most state data organizations really don't pick up a claim from a payer and they don't normally pick up a claim midstream. Essentially, hospitals and other providers simply have an electronic copy then that is modified slightly. If, in fact, there were requirements for a claims attachment that wasn't electronic -- and, of course, we would really only support electronic claims attachments -- it would seem to me that throughout the development of the multiple claims attachments, theorems that you are coming up with, the cost for development of those would be borne essentially by those that are going to be passing the information on.
So, I would think that there would be some involvement by payers. I think hospitals, obviously, would have some involvement, too. Most states that require this information do not pay for it separately from hospitals or from payers. I can't certainly speak today to say that they would be willing to bear any additional costs on that.
MR. MOORE: But as a -- if it is a free good and I am the ultimate user of that free good and someone else is paying for the collection of it and the development of it, et cetera, it becomes, you know -- I could see creative, innovative people saying, well, I could use this, I could use that. If it is the research community or others who were downstream in the process that want to analyze it, like NAHDO or other organizations, such as that, and I think some consideration has to be built into what are the real uses. Because I know in HCFA, we really try to go after our policy people and our operations people, who are never satisfied with the data that the claims people collect in order to adjudicate the claim. And, you know, they always want more and we are not providing enough in the operational setting. And to open that up to the entire universe of potential users is kind of frightening to me if I have got to pay for the collection and the movement of that data ultimately to the source.
I don't know where the providers stand on that particular --
MR. LUNDBERG: There is a real balance, as I mentioned; the burden of collection versus the value of the information. We wouldn't see -- 400 elements, you know, would -- may be a small core data set of additional information that would be designed not to add a lot of burden to anyone along the line.
MR. DAVIS: I would like to reply to Bob's comment, basically just give some experience of what we have gone through in New York State in the last five years, since UB-92 became involved.
New York State, as many of you know, used to have a proprietary billing form called UBF-1. We are no longer collecting the UBF-1 because it was a great burden on providers to providing a UB-92 and a state billing form. So, when we change, we, the state organization who collects discharge data, changed to go to the UB-92, we convened a group called -- what we called the Universal Data Set Task Force, which was a group -- a cross section of all our payers in New York State.
We have five Blue Cross plans. We have our Medicare/Medicaid, the HMO coalition, along with provider organizations, to decide what would be a common way to collect this element. Right now, the instrument that the SPARCS system uses to collect the data to try to be as minimal burden as possible on the provider community is the same instrument. It is the UB-92 flat file format, you know, Version 4. So that that same format can be used to adjudicate a claim and also send to us.
The way we made it work in New York State is our migration path -- and that is one of the reasons why I am here today is that we defined some state filler to deal with our additional data needs within that same format. Therefore, we decided in New York State if we are going to collect this data, we needed to collect it the same way it was going to be billed for the elements that were common.
Like Michael said, most of the elements that were collected are definitely common. It is just a matter of getting people to agree on the same definition. We have taken that step, but we still have additional data needs that are never going to be met totally by that claim, the need to be taken in that context. So, that is where we have to go.
I also want to talk about another question you have in terms of one of the -- with all of this goes a responsibility on both parts. There is a responsibility, I think, of the standards organizations to meet legitimate needs and there is also a responsibility on the data collectors to accept their needs.
For instance, in New York State, we used to have a reimbursement law that covered all payers in New York State. We used to set the rates for all payers. That reimbursement law changed. The data used to set those rates was, guess what, was our data. When that reimbursement law changed, one of my first steps was to go to those reimbursement people and say you now have new requirements. Do you still have the same data requirements? The answer was obvious. It was "no."
So, this year, when we announced our changes to the provider community, it was four elements that are no longer needed because we now have a change in the guidelines. So, I think that that is the responsibility of the state data collectors, to decide what they really need, but then we also national organizations to recognize that we do have needs that need to be met.
DR. COHN: This is an interesting discussion because it does speak of how, you know, we all want the same data except we all sort of don't really need all the same data.
Mr. Lundberg, you had mentioned the desires of the various states to come together to create some increased consistency in their data sets. I represent one very large managed care organization that in some cases, for example, has a site in Kansas City. And, as you know, the difference between Kansas and Missouri and Kansas City is a street. That defines the difference between the states.
Similarly, we also have other divisions that are in the Northeast comprising New York, Connecticut, Massachusetts. Now, in your vision of the future, do you see one state data set or do you see a format as a claims attachment for additional information of 50 versions, not unlike the UB-92 that now has 400 versions, somewhat based on the state filler issues?
I mean, what sort of consistency do you think we can get to, recognizing that many of these data elements have -- are part of the regulatory and probably even legal climates from state to state? Let's be realistic for a second.
MR. LUNDBERG: Did you say let's be realistic or did you say let's have goals? Let me answer both those questions.
The first goal, of course, is to try to come together to design a standard data set that includes both the standard collection document, currently, the UB-92 and potentially any claims attachment that would address all those needs.
The reality is that there will be some states that will not be happy with that. To the extent that they are still required to conform to those, I think that people will make adjustments wherever they can. There is this other larger issue of -- well, not larger issue -- of whether not state data organizations or others that are not engaged in transactions, but collecting this information, are specifically excluded from that.
I would like to be able to say that we receive response and we know exactly that. We are not positive about that. If, in fact, that were the case and to the certain extent these organizations would have a by and still be able to collect that, then there is no question in my mind that the UB-92 and the ending claims attachment could be tooled to meet everyone's needs because there would still be this other option.
But that is a question that has not been answered yet.
MR. DAVIS: Yes, I would love to respond.
Realistically, you have 50 state legislatures and it is very difficult to control 50 state legislatures. I know in New York State we have a very active legislature. We don't always pass our budget on time, but we do have a very active legislature and there are always going to be having requirements based on that.
So, that is your number one. I also think -- when I talked about responsibility on both sides, when we went from the UBF-1 to the UB-92, the biggest beneficiary has been the quality of the data that I collect. Once we went to the standard in terms of the way the hospital collects the data for payment, we realize that the quality of our data improved dramatically.
I will give a few examples of that. I testified a couple of years ago before the National Committee that we were collecting a unique ID and since there was no national standard on the unique ID, our legislation for the fetal death certificate or for the death certificate decided that we needed to come up with a New York State, a unique ID. We did. That unique ID is something at this point that is pretty much used internally. We have been able to take our birth and death vital records files and compare it to our discharge data files and we can just start deciding on data quality decisions without having to go to the hospitals to decide that we have another audit because we have a second file.
So, we have been able to decide problems with the vital records files versus problems with our internal files. And what we are seeing is because we have made that movement to standards, the quality of our data has improved. All right? And that is why you see us here today because we know that it is in our best interest to try to make that data set as small as possible, the differences as small as possible.
Will there always be differences? I maintain I don't think we can avoid that.
DR. LUMPKIN: Jeff.
MR. BLAIR: Help me understand a little better. From what I understood of what you were saying, both Bob and Michael, it sounded as if what you needed was standardization of the data sets that would be in a claim and I didn't quite understand what you needed in terms of a claims attachment.
MR. DAVIS: In the case of our claims attachment, okay -- and I will beg to my definition of "attachment," which is additional information. It is information that at this point is not on the claim. When you actually look at the numbers, we are probably talking -- we are talking probably four or five fields. We collect information now on alternate level care that is not on a claim. I don't know where else it is. So, it is, like I say, we need a home somewhere for it.
MR. BLAIR: So, you would like that to be included but as far as you are concerned, if they could be included in the claim, that would meet your needs just as well as if they are in the attachment or it is not appropriate to be in a claim or --
MR. DAVIS: Our needs are is that we need to go to a national standard or the standards development organizations and find a home for our place. If the appropriate home is in the claim, so be it. If the appropriate home is in the attachment transaction, so be it. Because if, you know, the technology is here for us to be able to link the two together, we don't want to be different. We want to be able to get the data the best way possible.
MR. BLAIR: Where do you feel is the appropriate place for this to be?
MR. DAVIS: I think it is going to be element by element where it is appropriate. You know, some things, as in my examples, would fit into the UB-92 coding structure and that is the appropriate structure, which means it would fit into the claim on the institutional side, the claim standard, all right, because that is an external coalition and would be run fine.
In the case of the diagnosis indicator, as I said, that is clearly attachment information, clearly not claim oriented information, but it is not logical to move it outside of that. We collect a method of anesthesia for us. Where that belongs, I don't know. I mean, if someone told me that they felt it was appropriate in some of the attachments, that is fine.
I guess the other thing I come with attachments -- and this is a little bit a part of SPARCS, but thinking a little bit globally -- is we, the department of health, like all other data, you know, organizations, like ourselves, have these things that we call registries. You know, these are these little, you know, pockets of people within the department of health. They collect information. Currently, we have a cancer registry, which is a national registry. Currently, we have Alzheimer's disease. We have congenital malformations. We have AIDS registries.
Much of the information that they collect, just like Bob said, it is very duplicative of what is already out there. It would be better if we could say to them -- and we have actually made this -- in this year, calendar year 1998, we, SPARCS, have gone to the congenital malformation people and the Alzheimer's people and gotten agreement with them to say you use the common elements -- well, we don't collect name, but zip code, county, you know, all the demographic information. You use that. You comply to the national standard, which we are using on the UB-92.
Then if you have additional data needs, which they do because they want modifiers on their diagnosis codes. Alzheimer's disease wants next of kin information because, obviously, they don't necessarily get that information from the patient. So, we need places for those. If those places are in an attachment, so be it.
MR. BLAIR: The other question that I had, Bob, a couple of times you mentioned SDOs, which I always translate in my mind to standard development organizations and I thought you were --
MR. DAVIS: That is what I mean, too.
MR. BLAIR: I thought you were implying that there needed to be some coordination with SDOs and I didn't know what you were getting at.
MR. DAVIS: Actually, I have been going to ANSI for a little while and I have been going to the National Uniform Billing Committee meetings for a long time because those are recognized standard development organizations. What has typically been the experience that I have had at those organizations is that they are just beginning to become aware that there are additional needs beyond the claim.
I think that the coordination is that they need to take that global view to understand that there are needs beyond just the claim and that it is the provider information systems that we want to simplify.
DR. MC DONALD: Part of my question was answered. I was trying to figure out whether what you are talking about is bigger than a bread box or bigger than a universe, you know, in terms of the additional data, what you are talking about. I guess you sort of said it, but you said it both ways.
You said there were four other fields. So, that is smaller than a bread box. Then you said everything else, all clinical data, at least that is how I heard the five or six disease registries you talked about --
MR. DAVIS: Yes, but, again, the four that I was talking about, Clem, was -- or Dr. McDonald -- is my system for --
DR. MC DONALD: Clem is all right.
MR. DAVIS: Okay. Good. I have heard your name so much at all these meetings.
PARTICIPANT: You can call him Dr. Clem.
MR. DAVIS: When I talk about the four, which I did, I was really, you know, narrowing my focus just to my SPARCS system, but we are beginning to understand that that focus, you know, just like everybody else, has to widen. I think that that was Bob's point.
DR. MC DONALD: The point is if you -- well, the question really is is if you are going to take and put the whole medical record on the attachments, you will put us out of business for the next two and a half years.
MR. DAVIS: I am not trying to do that.
DR. MC DONALD: Because the committee is charged to deal with medical record data in general in another six months. I think, in fact, that jamming the whole medical record into an attachment into sort of claims orientation might be a disservice to the data and to the providers.
MR. DAVIS: Actually, we are doing exactly the opposite in terms of what we have already done, the work that we have done with the genital malformation people and the Alzheimer's people is we have told them they can't have things. I mean, we have tried to --
DR. MC DONALD: No, no, I am not saying they shouldn't have it at all. I think will have it --
MR. DAVIS: They have to justify it.
DR. MC DONALD: Well, maybe I am not -- the bill says some other stuff in it, the HIPAA legislation that we haven't gotten to yet. Are you aware of that?
MR. DAVIS: Yes.
DR. MC DONALD: It is called electronic medical record. So, they don't have to be constrained in what they want if they do this right. If you have to fit it on one claim form, you have got a little bit of limitation. I don't think -- I mean, I think the burden of claim process with all clinical transactions is going to make it all break.
MR. DAVIS: Yes, but I guess what I am saying is I know about the process to know that the way the 275 has been built is to do an electronic staple. So, if we can do an electronic staple, then we don't have to burden it with the claims process because we can eradicate it out of that process and decide what is additional is additional.
I mean, that is what we are hoping will work. That is what I have been working for through ANSI organization. I believe it will work in terms of the way that transaction site is set up. That is what we are working for.
Yes?
DR. LUMPKIN: Let me, since I was next, kind of walk through this process because certainly public health information is kind of near and dear to my heart and having started and having the trauma registry in Illinois being my baby, having started that one from scratch, I have a lot of interest in that. But what I am having some difficulty understanding is that the information that you get, even though it is in the format of the UB-82 --
MR. DAVIS: 92.
DR. LUMPKIN: I am sorry -- UB-92. I am living in the past. I find myself doing that more and more.
But the hospital when they discharge a patient, they send the UB-92 to their payer in whichever flavor their payer requires. Then they send a duplicate copy to you. So, it doesn't go to a clearinghouse and get stripped off and sent to you. You are actually requiring under your state law information, which, in fact, tends to mirror the UB-92.
MR. DAVIS: The state law says that they do things according to our specs. We have decided that the quality of our data is better if our specs are very consistent with the payer specs. So, the hospitals in some cases do exactly what you just described. In some cases, they have decided that they want to strip off the data and do separately. That is totally the hospital's choice.
DR. LUMPKIN: So, the kind of information that you are requiring and requesting is not -- maybe you can help me here. It doesn't appear to me to be covered by HIPAA.
MR. DAVIS: That is a question I talked to Mr. Braithwaite about. I don't know if it is or not, but I will go back to, you know, whether it is a legal responsibility or a moral responsibility. This is kind of how I have talked of this with my staff is that if I decide -- let's say I am not covered by HIPAA. Let's say, you know, no state discharge system is covered by HIPAA and we decide to go off in our separate direction because we have our own state legislative mandates, I may be able to get away with that, but I will maintain that the quality of my data will go down hill.
One of the things that happened, without a doubt, is the quality of our data improved when we went to UB-92. I mean, there is no question about it. So, I probably would be shooting myself in the foot if I did that.
DR. LUMPKIN: And just follow me down this path a little bit further. If, in fact, what you are not -- you are not receiving a claim. You are receiving information that has been formatted in a different stream to be a claim and now is being formatted to be a report to the state.
MR. DAVIS: Yes. That stream is exactly the same stream that they send to the payers.
DR. LUMPKIN: Right. And given the discussions that we have had earlier about -- because one of the things our committee is trying to do is get our handle around attachments -- I am not clear on why you perceive to want to tie into the attachments as being the source of information since the attachment seems to be designed specifically to answer questions, which we don't fully understand how the -- how or why the payers ask for it. I mean, that is something that we need to define if we are going to standardize -- why you see that as being an area that is important to standardize for state data organizations.
MR. DAVIS: I guess the answer to that is my strong belief that the only way that my data will stay good is if it is attached to a national standard. I know that the claim standard can't support every need that I am going to ever have or every need that my legislature is ever going to have.
So, knowing that I need a home and that was the context of my statement is I need a home for my stuff. Quite frankly, I don't care where it is. I want a logical home for the data that I can collect, the data that I need. I mean, I don't want to just go off and get frivolous data in answer to Clem's thing. I want hospitals to justify it. We have done analysis -- you know, in terms of answering your question, we have done some analysis in going back to people who have asked for data that was frivolous, we have estimated in New York State, because we collected 2.5 million discharges a year and then we collect another million or so ambulatory surgeries a year, that for the one character of information -- and this is not putting it on the screen, not for the vendor to pay for it to be put on the screen, just to pay the clerk to enter the data, look it up in the medical records, costs about a half a million dollars a year, just dealing with those expenses.
So, with a number like that that you can get your hands on, half a million dollars year, can the people who are using that data justify it. If they can, fine. Where do you justify it? Do you justify it in our legislation? Do you justify it by some cost savings somewhere else?
Those are the kinds of questions that is our responsibility before we go to national data organizations. But once we have done our homework, once we have done that, then we feel we need a place for our data. All right? Not to overburden the claim, but to find a home for it.
If you maintain that the attachment transaction set, the 275, which is -- I am thinking it will because that is where I have been spending a lot of time -- will do the trick and I do believe the 275 will do the trick, that is great. If you tell me that it is not going to do the trick, then I would ask for some guidance as to where to go.
DR. LUMPKIN: Well, two follow-up questions on that because if the ball park that we were given -- somewhere between 3 and 20 percent was the testimony we got earlier -- of claims require or generate a claims attachment and as I understand the function of your -- of the hospital discharge databases, they really require every discharge to generate data.
Then would you perceive if the claims attachment was used as a vehicle, therefore, requiring a claims attachment on his hospital discharge --
MR. DAVIS: Claims attachment for what we would need, yes.
DR. LUMPKIN: Then I guess my last question then following down that line, you mentioned 275. If, in fact, 275, as I understand it, is just the envelope, would you see a need to, therefore, go on further to define some content to that envelope that would --
MR. DAVIS: Absolutely.
DR. LUMPKIN: -- give some structured and standardized format to the information you need?
MR. DAVIS: Absolutely. That is why we are working through the ANSI process and the Proof of Concept Team to try to define what are additional data needs to put them into the spreadsheets, to try to define what that content is. That is why we make appeals to other national standards developments organizations to help us include us. Several of our data needs could be met with the coding structure for NUBC and we have, you know -- George and I have been talking for a long time about those needs and how they could be best met.
DR. LUMPKIN: And then finally, we understand -- I think the latest number is about 30 percent of all claims are submitted electronically. Is that a number I heard recently?
MR. DAVIS: Yes.
DR. LUMPKIN: What is your experience with electronic submission of hospital discharge?
MR. DAVIS: A hundred percent.
MR. LUNDBERG: A hundred percent.
DR. LUMPKIN: Okay. So, we are talking an environment where it is already automated. Thank you.
DR. COHN: I was just going to ask sort of a follow-up question to what John was asking, which is that, obviously, you are talking about a claims attachment with certain data elements and certain code sets in those data elements and all that.
Who in your view would be responsible for the upkeep and maintenance of that particular piece? How would that -- I mean, that whole area is sort of a yet to be determined sort of across the continuum and how -- how do you see that happening?
MR. LUNDBERG: If, in fact, it is the purview of this subcommittee to design those claims attachments, then if it is the purview of that, I would think that periodically that information would be reviewed for the adequacy and whether or not it is being used and, of course, it should be tossed if it is not being used.
When you talk about maintenance, we have really talked about passing information on. I know many states would do the editing at their own level and interact with hospitals if there were errors, as they do now.
DR. COHN: No, actually, what I mean is maintenance of the data set and the standard as opposed to the -- you know.
MR. DAVIS: Yes. Actually, and I begged to Marjorie, this is one of the most encouraging things that has happened, you know, in the last several -- well, actually, several years -- the fact that the National Center for Health Statistics brought together a group of people with similar needs and will be organizing some way to get these people to begin the process and the process to begin talking about what the common needs are.
So, I am seeing just like, you know, before HIPAA, we had fragmented groups in terms of what the standards would be across the country. That was certainly true with, you know, state data organizations like ourselves, who really just did their own thing.
And then we are coming to find out now through the work of NAHDO and other groups is that we do have common interests. So, that needs to go farther and New York State will support that effort a hundred percent and they should be commended.
DR. LUMPKIN: Jeff.
MR. BLAIR: Could you help us understand how far along NAHDO is in driving towards common data sets for all states? Are we a year away from that or two years away? Is that a stated goal? Are they actively working to do that? Where do we stand on that?
MR. LUNDBERG: A little over a year ago, they polled all state data organizations to understand exactly what they were all collecting and they have come out with a core set of information that everyone has collected and then a frequency count of those other elements that some collect and then sum those.
They have now designed a questionnaire to tease out further information on what their real needs are and from that. I believe that it would be less than a year to which they would have gone through the discussions with the states to find out what states minimum and actual requirements are on this. I am not saying that they would necessarily have consensus. We haven't explored it in that fashion, but NAHDO is well along the road of understanding what they collect and is now fine tuning what is a drop dead position on states. And that would be the process that would be the most difficult.
MR. BLAIR: Did the core data set seem to cover 80, 90 percent, a very high percentage or is --
MR. LUNDBERG: A very high percentage. I mean, if you look at it, most states are basing their data collection efforts around the UB-92. Many states have then gone to standardized information as of a single physician identifier, a patient identifier, which is not there on the UB-92, and some other elements.
The other things are additional clinical information that are not medical record-based -- that are not a medical record, but another six to ten elements, four to eight, depending upon who you speak with, that address needs of researchers and of the ability to adjust data for differences in severity.
So, it is not a huge -- I don't believe there is a huge additional amount of information that people are collecting. There are one or two states that collect more than others.
MR. BLAIR: Let me go see if I could -- if there is another linkage here. Bob, you had mentioned coordination with SDOs. Are the data sets that you have identified in that core data set, do they link back to code sets where standards are available within provider institutions, things where you could pick up a LOINC code, a SNOMED code or -- are the codes there for you to be able -- for us to be able to automate, to feed in?
MR. DAVIS: Most of the codes that we collect, and I will agree with Mike at this point, we don't need to go to a SNOMED or LOINC at this point for the discharge stuff because it is already a part of the UB set, which is already part of the standard because most of that -- really, a large percentage of it is already there in terms of not having to go to attachment. It is part of that claim transaction.
We are exploring that, though, because that -- I mean, that is part of what I have been doing with the Proof of Concept in the 275.
DR. MC DONALD: A follow-on --
MR. LUNDBERG: You were next. So, you can even change the subject if you want.
DR. MC DONALD: No, no, no. It is actually -- I am trying to get a sense of what these really are.
Mike, you have got the -- could we get a copy of this list of the data elements because it always to see what they really are?
MR. LUNDBERG: When you say the "list," -- I am sorry -- the list of Virginia data elements or those of all those?
DR. MC DONALD: You have got them all collected and you have got -- it would be very helpful to --
MR. LUNDBERG: Yes.
DR. MC DONALD: Then the follow-on is that you said you have got six clinical data things that aren't part of the medical record. Who makes those up then? Is that made up by the administrative staff?
MR. LUNDBERG: Generally, there are -- well, I can speak in our state. It was a combination of a discussion with providers --
DR. MC DONALD: No, what I mean is where are they found in the hospital, in the -- who makes them up?
MR. LUNDBERG: Well, essentially, it is all over the board. We collect information on patient race. That is something that is collected by the admissions clerk.
DR. MC DONALD: Well, those six things you said, these six clinical things that you said weren't part of the medical record. I am just trying to understand what you are saying because I can't picture that.
MR. LUNDBERG: Well, sure. One is E-code, the external cause of injury code. That is something that is picked up from the medical record. Another that one of the states is a "Do Not Resuscitate" order. That is picked up from the medical record.
DR. MC DONALD: They are medical record things.
MR. LUNDBERG: Some are, some are not. Patient's social security number, you know, you are going to get it from a demographic section.
MR. DAVIS: They really are just, you know --
DR. MC DONALD: The other thing is I got a -- I was, again, at one of the Proof Concept meetings in Chicago and one of the people there gave me a set of a questionnaire they had from public health. But it was like 20, 30 pages long. It had at least 500 questions on it.
DR. LUMPKIN: That is probably an Illinois record.
DR. MC DONALD: I mean, it was a very good form. It was an excellent form.
MR. DAVIS: I would like to see it. It was kind of like more questions -- I have got it in a stack I am supposed to read it at home.
DR. LUMPKIN: But that could be a cancer registry where there is clearly identified disease and more detailed records. That would not be the medical discharge or the hospital discharge data set.
MR. DAVIS: There is an interesting sidebar on that because the reason why the congenital malformation people and the Alzheimer's people have come to us to try to help them collect their data is that when the hospitals submit that data, they have these cards. There is a blue card and there is a yellow card. So, they know all the different colored cards -- Kathleen is shaking her head, but they have these cards and what is happening is as monies dry up all over the country, they don't have money to keypunch that data. So, they are now looking for alternate sources of ways to get it in.
So, now, when they came to me, the condition was we will provide an alternate source. We have the transportation vehicle, which is this Version 4, plus our state extensions, but if you are going to use our transportation vehicle, you have got to use our standards. So, they couldn't collect name in a different way. They couldn't collect sex in a different way.
They couldn't collect the diagnoses in a different way. That was the preconditions for them to use our vehicle.
DR. STARFIELD: You have both focused primarily on institutional -- information from institutions. We heard this morning that most of the activity surrounding the transactions is non-institutional. Do the states have any interest in that information, any thoughts about that?
MR. LUNDBERG: In Virginia, we collect information on state employees in Medicaid, all care that includes physician information, ambulatory surgical center and others. Most of the issues that we would have about standardization with those will apparently be taken care of with other aspects of the Health Insurance Portability Act for the payer ID and the provider ID number.
Those are the main standards that we are looking for. Sure, we could all benefit from more detailed and rich data, but at this point we don't think that this is something we would be working towards.
MR. DAVIS: A hundred percent agree with what Michael said. Whatever professional data would be collected in New York State would be through the New York State Medicaid program and, again, they are working through the ANSI process to deal with that.
I don't see anything on the horizon where our clinical data will be extended to the professional -- in fact, I don't even see it being extended to the whole world of outpatients because the numbers are so staggering at this point.
DR. LUMPKIN: I just would perhaps suggest that this may be an issue that the full subcommittee may want to take up because I do think it relates to the broader mission of the committee, not just our HIPAA responsibilities that we may want to look at some issues of standardization across the clinical data sets.
I know that there has been standardization within registries but not across registries. That is more than just want NAHDO does and other organizations. And it really calls into question what we are doing in public health and what we are requesting from the health care community and that is something that we may want to look at in a non-HIPAA format.
DR. MC DONALD: I don't think the impression we got this morning -- it was mostly doctors, but I work in a hospital and we also looked at the prevalence of rough requests for attachments and one of them was hospital discharge summary on the top 12. So, there is still an attachment traffic of significance in hospitals.
MR. MOORE: But I think when this group was formed back last summer to look at attachments, the decision was made that attachments is far more than just what goes into a payer because, as he said before, one of the reasons is to improve the system at the source of the data, which is back at the provider, regardless of who is the ultimate recipient of that data downstream.
So, in the payer community, yes, the attachments might be far more frequent on what would be the supplier, physician, lab, type of service, where you have the volume. Like in Medicare, it is 1.8 billion line items on all those Part B claims, but on the institutional side, home health and skilled nursing and hospital, although a lot of the money is there, the volume in claim services aren't there. So, therefore, you have fewer items.
For example, we get 130, 140 million Part A claims, all institutional claims per year. And some of you out there if you know better -- but we get about 700 million Part B claims. So, the volume is in that side. Hence, the more attachments.
DR. LUMPKIN: Other questions?
[There was no response.]
Thank you.
We are going to have our lunch break. We are scheduled to come back at 1:30. The committee is going to reconvene at 1 o'clock and we will have some general discussion amongst the committee and then we will go -- when our panel shows up, we will launch into the panel.
So, we will reconvene at 1 o'clock.
[Whereupon, at 11:53 a.m., the meeting was recessed, to reconvene at 1:06 p.m., the same day, Monday, February 9, 1998.]
DR. STARFIELD: We will start the panel for this afternoon as soon as we have all the testifiers. So, I think from now until that time, we will just have open discussion, continuation of questions that we had before and anybody is free to ask a question and to answer the question and continue on with anything that we started this morning.
DR. COHN: This is just open discussion?
DR. STARFIELD: Yes.
DR. COHN: I guess I find myself having almost having almost more questions not with necessarily answers, but just sort of -- obviously, as I am listening to the conversations, what I hear is that, you know, people have data needs, have needs that have been, at least, heretofore, met, at least partially by the UB-92.
There is, obviously, ongoing discussion or innuendo about stripping out a lot of that data from the 837 transaction and putting it into something called claims attachments. Now, this is all, I think, what I am hearing, but nobody has ever said that directly, as well as, obviously, as one reduces the variation, there needs to be another place to put all that.
Obviously, I find myself just a little concerned to make sure that whatever happens, happens in a way that is sort of seamless and easy for the industry. I find that this is an area of concern.
John, you look like you have a worried look on your face.
DR. STARFIELD: Do you want to answer that, John, or do you want to let Jeff give you time to think?
DR. LUMPKIN: Well, Jeff has his hand up and I don't.
MR. BLAIR: I think that there is a -- is Clem back here because I --
DR. STARFIELD: Clem is not here yet.
MR. BLAIR: I think that there is a problem with different sets of expectations for what electronic claims attachments should be yielding us, what we should be getting from that. Because when I heard the testimony this morning from the payers, their expectation in my mind is very, very low. All they expect to be able to get out of this is reduction in clerical work; in other words, getting information faster, being able to ask questions and process it cheaper.
There was none of the testimony that I heard, unless I missed something, where they are anticipating this is going to reduce their evaluation processes. I think that that is okay and I think that there is validity from a short term perspective. And if that is the case, then maybe we -- if it turns out that we continue to get testimony that is consistent with that we got this morning, then maybe we should focus on near term, short term envelopes for health claim attachments.
And I think the problem is that there has been a number of us that had much higher expectations for where this could go to because we were beginning to think that folks could prioritize, could focus, where we could focus on particular questions and codes, get them automated, where you could begin to improve business processes as a result of this. And I think maybe we ought to just, you know, put that as a long term objective and then if we do put it as a long term objective, then maybe that also changes.
Because if we get to the point where we have clinically specific electronic health records, a lot of the process improvement will be clinical process improvement that the providers will have because they have to compete with other providers and the policing that the payers are going through will diminish. The need for that will diminish.
So, the roles are going to change from a long term perspective. So, this is not a final answer or final judgment. This is just sort of -- I am beginning to -- my thinking is beginning to evolve into this long term, short term view and if it -- and the pieces that are long term, I am beginning to think the roles are going to change in the long term. So, maybe we should just focus on a short term claim attachment answer and not try to bite off more than we could chew.
Thoughts? I don't see -- I can't see whether faces are grimacing or agreeing or what.
DR. STARFIELD: Well, we do all agree that short term is relatively short term because on six months we have to be on the long term; that is, the electronic medical record, within six months. So, whatever short term we do is very short term.
DR. COHN: Well, but -- actually, let me just make a comment. I may be the only one who interprets HIPAA the way I do, but as I look at, what is it, Subtitle F of HIPAA, what I see is is that there is a statutory agreement that claims attachments and other administrative transaction standards need to be implemented.
I see that the area around electronic medical record standards to be more on the level of advice to the Secretary. Am I the only one who interprets --
MR. BLAIR: I agree with you, Simon.
DR. COHN: -- the legislation that way?
MR. BLAIR: Yes, because I think that most of the things that people expect from electronic medical records are long term solutions. They are not some things we are going to get in a year or two.
DR. COHN: Yes. It is just that I think that we had better make sure that our short terms are extensible because I don't think we can guarantee necessarily that the other piece will fall in a predictable fashion.
MR. MOORE: I think what the claims attachments -- this is some of the things that -- and I have heard Clem mention this before and he said it this morning when he was talking about things that are in the claim today that need to be taken out. And he mentioned things like the hematocrit level, EPO, things of that nature, which HCFA has put into the claim, but I think as you move into this claims attachment area, you are finding the clinical information that payers are going to use on making future decisions, not current decisions.
I will give you an example at HCFA, where we last year -- and the EPO and hematocrit level are part of our claim -- and last year, we made a policy decision that said we would not pay for EPO if the patient's hematocrit level was 36.5 or above. That meant we wouldn't pay for any EPO that month.
So, docs immediately began to change some of their practice and the EPO level went way down on these patients. Well, we are really getting our comeuppance now. And, so, we are back to the table to look at our policy. And in looking at the policy -- to make it based with evidence not subjective evaluation, we are using that same data, like with a hundred percent of the patients to look at what happens to a patient when you take away the EPO and do you take it away incrementally, et cetera. But this will affect future payment policy that we are going to try to change.
So, it begins to look at this data and I think the states will be using that data in a similar fashion and a lot of organizations. So, it is not just a payment decision that is based on that particular claim, but a lot of this information begins to formulate the very reimbursement policy for future care and quality issues that we are all beginning to face.
So, I don't know how you factor that in.
DR. COHN: Can I just ask you a very concrete question?
MR. MOORE: Yes.
DR. COHN: This has to do with just as an example, this EPO piece. That is included currently in, I know, the UB-92 pieces. Is that included in the 837 in any, way, shape or form?
MR. MOORE: Yes.
DR. COHN: Okay. Now, speaking of the industry -- now, just go along with me on this one -- we are acknowledging standards for the 837 to go into effect sometime in the year 2000 and about a year later, the claims attachment standard will be implemented. So, that means that initially because there is a need to capture the information described around EPO, that will continue to be part of the 837 standard and then we will gradually or in the next year be somehow taken -- moved into the claims transactions -- excuse me -- the claims attachment piece.
Is that going to cause problems in your view, I mean, that sort of changes in the standard from year to year?
MR. MOORE: I didn't hear that it was going to be taken out of the claim.
DR. COHN: I am just hypothesizing.
MR. MOORE: I don't think we have got a decision from the group yet on that. There is more than just the EPO that is in that particular claim that has been stated to be clinical information. If it is, then I would assume that it is like the Certificate of Medical Necessity that is in another piece of documentation that is part of the claim.
The reason I think a lot of things have been added to the claim is there was never any other place to put it. So, there was no attachment in the past, other than the photocopies of clinical records. Just with this, I think, initiative have we begun to think more broadly about what are the different transaction sets and how can they be worked together to form an entire picture of what is happening in the health care delivery for that person.
DR. COHN: Yes. And I guess I should just clarify that I was hypothesizing as opposed to making any definite statement about that.
DR. STARFIELD: Are you ready?
DR. LUMPKIN: Yes, I think I am ready.
I think my difficulty in dealing with this is that I don't get the feeling we are conceptualizing the big picture. We have a system or a business of providing clinical services, healing services, caring services, treatment services to individuals. That is the nature of it.
Now, in order to be able to do that, we also have to have a way to pay for it. So, our system has evolved where, by and large, the bulk of the automation has occurred. It has occurred in the financial end. Now, if you are in the business of making beepers, like mine that just went off, you say your business process is how you make that beeper and that your support systems are how you bill your customers.
Our business processes in the health care is to provide these services. There is a whole bunch of information that is done when a patient arrives in a clinician's office. One individual collects the financial and the business information, the fiscal information and generally the clinician collects the medical information and it requires some abstraction.
So, as we look at this concept that is an attachment, it really is a way of trying to get clinical information to make a financial decision, not to make a clinical decision.
Some of the texts that I have seen refer infrequently when you set up business rules to a 20 percent rule, that unless it occurs more than 20 percent of the time, you handle that specially and that you set up your systems to handle those things that are handled in the vast majority of the instances. By and large, that is what our transaction forms are.
But until we came up with the concept of attachments, we would just add more to the billing form. So, in fact, we do have things in 837 that don't pass the 20 percent rule. If one were going to clean it up, then that would be one of the things that would be done.
I think we have to recognize, though, the totality of our task and that is that as we are trying to move forward, we may be moving much further ahead than the technology of abstracting information from basically a paper system even today, which is clinical information, into the financial system.
So, we call that a transaction -- I mean, an attachment. And it is almost as if by coming up with that word we have expected it to define itself. And what we really ought to be looking at is the function. Some of the function of transcribing information from the clinical record occurs under 837. For those that don't, rather than trying to cram all this stuff into 837, we are going to now set up a new mechanism.
That is my perception of what we are trying to do with the transaction, is this new mechanism for all the stuff that isn't somewhere else but is not required on every single bill. So, it is going to have to be a menu. And I forget the name of that restaurant that we went to last time at the station, the one that had the menu with like a thousand things on the menu, the America.
So, it is not going to be a limited menu thing. I think it is going to be a menu that we are just going to have to continue to add things as we get into agreement of different clinical information but that perhaps our challenge as a committee is to at least get the menu started and to pick off those things which are high frequency attachments, have some agreement on how they should be formatted and then begin to add to that as time goes on.
MR. MOORE: Since we require this hematocrit and EPO data on every claim for a dialysis patient, is that something that this committee would say stays with the claim? Or, you know, we require a Certificate of Medical Necessity for -- we, HCFA -- for every DME claim? Is that something that stays with the claim? Or do you have the latitude to say that those who absolutely require that data in order to pay a claim pay the price for it with those providers who provide that service? Where other payers and other people in the community do not need it or do not want it, do they have the option of not getting it?
DR. COHN: I hear your question. It is not one that I would choose to answer now. I am hoping that there will be additional information over the next day, day and a half, that maybe we will --
MR. MOORE: This is part of tomorrow's discussion.
DR. COHN: Well, no, but I think it is a question -- it is probably beyond that, but it is really, I think, where the rubber begins to meet the road in all of this is really how we implement this in a way where it really works and that the road through the transition isn't too bumpy.
MR. MOORE: And keep in mind as we go through this decision-making, as we move more into the quality of care issues, we need information like that to make those judgments on changes in policy as to what we are going to do with certain situations. That kind of information is not in the claim and almost in some cases if you are beginning to look at the managed care setting, you need to have it in order to evaluate what is happening.
DR. STARFIELD: Well, Simon, you made the distinction between what is required of the committee in terms of the HIPAA and the electronic medical record, which is only advice, but when you begin to hear that a lot of the claims attachments are, in fact, in the form of a medical record -- we don't know what proportion, but a lot of them for sure, then you begin to think maybe there is not a big distinction between these two tasks that the committee has and it is how we construe this.
Maybe we just need more information about how often the medical record is not only requested but submitted actually because more is submitted than is requested and, in fact, how much of that is used that we begin to construe our work more broadly or more narrowly, depending on that.
Well, are you ready for the panel? Shall we start with that now?
Okay. I thank this panel for coming ahead of time. I am sure this is going to be a very interesting -- now, we have heard a lot of hearsay about providers this morning and now we hear the real scoop. So, what I would like to ask you first is to introduce yourselves. Why don't we go that way and then we will come back to the beginning and do the testimonies.
From Dr. Cheng, introduce yourself and let everybody introduce themselves.
DR. CHENG: My name is Paul Cheng. I am the director of patient accounts and managed care at St. Clare's Hospital in New York City.
MS. LAUER: I am Elise Lauer. I am the director of patient accounting at Northwestern Memorial Hospital in Chicago.
MR. SMITH: My name is Dan Smith and I am president of Quantum Medical Business Service in Roanoke, Virginia and also the treasurer of the Billing Association.
MR. OWENS: My name is Bob Owens. I am the chief information officer of the American Dental Association. And my apologies, our president-elect was caught in the fog-in in Chicago, so he could not make it today.
DR. STARFIELD: Okay. You might want to reverse your sign and put your name on that.
Jerry.
MR. HENRY: I am Jerry Henry. I am chairman of the State Uniform Billing Committee in Kentucky and I work for Baptist Health Care, who have a network of hospitals across the State of Kentucky.
MS. NARCISI: I am Jean Narcisi and I am director of the Office of Electronic Medical Systems for the American Medical Association.
DR. STARFIELD: Okay. Paul.
DR. CHENG: Thank you.
Distinguished members of the National Committee on Vital and Health Statistics and guests, thank you for allowing me to share my views from a provider's perspective on a very important aspect of the health care claims process, claims attachments.
My views will be from both the institutional and medical practice perspective.
My name is Paul Cheng and I am the director of patient accounts and managed care at St. Clare's Hospital and Health Center in New York City, New York. St. Clare's Hospital is a 200 bed, HIV designated facility located on the west side of New York City, providing a wide range of HIV and non-HIV medical services, both on site and at several health centers throughout the New York City area.
St. Clare's Hospital is not a large facility like its neighbors, but St. Clare's Hospital does generate a considerable number of claims for both institutional and professional services. Due to the nature of the services rendered, the attachment requirements or, as I call them, problems, are considerable for both institutional and professional claims.
The daily requests by payers for additional information before a claim is to be processed for payment are increasing. Progress notes, nursing care notes, physician's consultation reports, plan of care notes, therapy notes and et cetera, are constantly being requested. In addition, once the primary payer has paid their portion of the claims, the secondary carrier in most instances requests the same information, plus a copy of the explanation of benefits from the primary carrier.
Recently, more and more requests for additional information by independent utilization review agencies representing payers are being made mostly in the form of the complete patient medical record. Should the provider fail to provide such information, payment will be indefinitely delayed.
At the medical practice level, the same phenomena is occurring, where there is an increasing number of requests being made for the physicians' records and a completion of more attachment forms. The physicians are complaining and they are complaining vehemently that the completion of the additional forms and attachments require them to re-review charts several times and then their office records several times, which takes away from patient care and adds additional costs to them.
These requests are costly and time-consuming for all of the providers. In most instances, providers are not able to determine why the documents or additional forms completion are required.
It has reached the point where the provider will automatically send attachment information if the provider has had experience with the payer, who has requested information in the past. Any information over and above what is requested on a claim form is assumed to be an attachment by the provider.
At least with Medicare and Medicaid carriers, explanations are given. Currently, managed care and ERISA-type programs have taken a lead in attachment requests and requirements. Many providers feel that these requests are for the soul purpose to delay payment, to protect profits, especially when non-clinical information is requested; for example, accident information, additional insurance information, family information, demographic information.
In my 25 years of patient accounting, changes and standardization in claims processing has been painfully slow. Resistance to change and the cost of change has been enormous and chaotic. Having been involved in both state and national billing committees for many years, I have seen both payers and providers alike argue excessively with each other in the standardization of the claim form.
Throw in regulatory agencies, advocate groups and other parties into the mix and you have a real mess. At least with the current UB-92 and the HCFA 1500, there is some standardization and simplification. The standardization of the attachments process will require significant amounts of tradeoffs, compromise, agreement and, most of all, trust.
Unlike other states, New York has an electronic claims submission requirement for payers. The national standards format is used for professional claims submissions and the HCFA EMC Version 4.1 is used for institutional claims. Most payers in the state are in compliance with the requirement.
In addition, the New York State Insurance Department has minimum data requirements for both providers and payers that are appropriate for claims processing. The New York State Universal Data Set Specification Task Force is composed of providers, payers, vendors, government agencies and other interested parties, which meet to determine the appropriateness of requests for data to be added or removed in the electronic data set.
Needless to say, providers and payers alike in New York State are ahead in the EMC world. The key to the success in New York State is the cooperative approach, mutual trust and legislative backing. I would suggest that when the attachment standards are adopted, New York State will mandate the same.
As a member of the New York State Uniform Data Set Specifications Task Force, I extend an invitation to you all to attend any future meetings to observe the process.
Once simplification and standardization is achieved in the claims processing and attachment process, I would suggest the following to be included: formulation of a single universal provider maximum data set for claims and attachments; providers, population of all locators. Payers glean only that information that they require for claims adjudication. Payers then forward the complete data set with their additional processed data to the secondary carrier. Appointment of a single body composed of a cross section of the industry, payers, providers and regulatory agencies.
And last, but not least, legislation be introduced and passed requiring all payers and providers to adopt the standards. There must be no exceptions or failure is inevitable.
The results would be a cost savings for providers and payers and programming, faster processing of claims, better information for multiple purposes, ease of reporting and on and on and on. Providers and, I suspect, payers would welcome the standardization of attachments, but be aware of one major sector of the industry, who will have difficulty; the medical practitioners.
They don't have the same experience nor have they made the same advances in electronic media claims as institutional providers. This sector will need additional time to reach the same level playing field as the rest of us.
My comments with regard to the questions are attached to my presentation. My thanks to the committee for allowing me to testify in this very important matter and if I can be of future assistance, please give a holler.
DR. STARFIELD: Thanks very much.
We will go on and then we will come back with questions later on.
So, Elise Lauer.
MS. LAUER: Thank you. Again, thank you to the committee members for allowing me to participate.
I am Elise Lauer and I am employed at Northwestern Memorial Hospital in Chicago as the director of patient accounting. I have been employed by Northwestern since 1975 and have served in various roles; the last 16 years, however, specifically in patient accounting.
Northwestern is a 750 acute care academic teaching facility serving approximately 270,000 patients last year and having processed --
DR. LUMPKIN: Does that include the new building?
MS. LAUER: I knew that was going to come up today. No, it is not open yet.
-- and having processed over 525,000 related claims. I am responsible for all billing and collection processes, including government programs, commercial and managed care payers, corporate, research and self-pay. I am also currently the chairman of the Committee on Patient Financial Services for the Metropolitan Chicago Healthcare Council.
This committee membership includes representation from approximately 60 Illinois hospitals and the group meets regularly to raise, discuss and resolve reimbursement issues impacting these institutions. I have served on this committee since 1993 and have been chairman since July of 1997.
Because Northwestern was one of the first institutions in Illinois, who successfully implemented and fully integrated the Medicare electronic remittance process into our day-to-day operations by developing a number of automated functions and audit tools as described in Attachment B, I was also asked to serve as chairman of the EDI Task Force. It is a subcommittee of the same group and I co-chair along with the EDI manager for the Health Care Services Corporation, who is currently the fiscal intermediary for the Medicare program in our area.
This group was instrumental in the successful implementation of the original Medicare EFT/ERA and continues to review and address issues relating to subsequent version implementations.
NMH has filed claims electronically for commercial, Medicare and Medicaid claims for several years and is able and willing to submit electronic claims to any payer who is able to accept them. Although approximately 85 percent of our claims are initially transmitted electronically, the increasing demand for additional documentation continues to draw upon our resources, increase cost and delay reimbursement.
It appears many claims attachments described, including those provided in my responses, can be categorized into three groups. The first represents requests for information that is already provided on the claim form. However, because a plan and/or payer is unable to interpret the data due to older technology/systems, which do not support current formats or the inability of the payer to keep current with the coding schemes. In both cases, the information is re-requested in a format suitable to the requestor's operation.
The second group represents information that payer's are unable to successfully obtain either directly from their employers, subscribers, healthcare professionals or other plans or payers. In these cases, hospitals' reimbursement is delayed and the hospital is charged with the tasks of gathering, formatting and submitting required information that in some cases may be unrelated to the services provided, in order to obtain reimbursement, like COB forms, authorizations, medical records in order to investigate or identify pre-existing conditions and benefits explanation forms.
These situations are the result of an industry unable and sometimes unwilling to effectively communicate with itself. As you may recall, in the early nineties, significant and unsuccessful efforts were exhausted attempting to develop local area CHINs to assist in resolving the problems associated with the overwhelming demand for access to information.
The third group represents information either supplied earlier by the provider or actually resides in the payer's systems. However, due to manual processing backlogs, inadequate storage and retrieval systems or just a genuine inability to link information, payers will request we provide the information again with the claim.
Examples would be the Medicaid split billing and spend down documentation required in Illinois, UR and case management documentation.
Another cause for increased manual processing is the wave of complex reimbursement arrangements and a genuine lack of products to support these arrangements. In order to maintain patient populations, providers are faced with increasing resources to accommodate the manual processes associated with administering these contracts.
As is the case with many of our specialty contracts, it appears the insurance industry is slowly transferring the costs associated with reporting and claims administration to the healthcare provider, a savings to one component but not a savings to the industry as a whole.
Overall, I welcome the legislation. It has the potential to provide definition and standardization and improve the mechanical functions associated with data collection, storage, transmission and processing. However, commercial payers have been slow to adopt standardization practices and to making their businesses electronically accessible.
In my opinion, the penalties described in the HIPAA legislation for non-compliance could, in fact, actually be the more cost effective approach to responding to the legislation, rather than developing or accessing the technologies necessary to support it. This is one case where I hope to be proven wrong.
With regards to the claims attachment, I would recommend developing mechanisms that would effectively evaluate the information being requested as a claim attachment, ensuring that to the extent possible, these data requirements are incorporated into the claims themselves.
Again, I thank the NCVHS subcommittee for the opportunity to participate. My responses to your questions are included on the attachments.
In addition, I also did provide the State of Illinois actual format, electronic formats, for information.
DR. STARFIELD: Thanks very much.
Dan Smith.
MR. SMITH: Good afternoon, again. My name is Dan Smith and I am president of Quantum Medical Business in Roanoke, Virginia and also treasurer of the Billing Association.
Today I have the honor of representing the International Billing Association. On behalf of the IBA, I would like to thank you very much for the opportunity to appear before you Subcommittee on Health Data Needs, Standards and Security. IBA is the only trade association representing third party billing companies. These companies employ nearly 20,000 people nationwide, processing an estimated 650 million claims per year, representing nearly 55 billion dollars in medical claims annually.
I have a brochure, which I will pass to the committee members, outlining what the IBA does. We are here today to explain what the third party billers' role is in processing claims and attachments and to answer questions that you have posed in your question outline.
Our primary business is accounts receivable management for doctors or physicians. However, a large number of our members do complete practice management for doctors. Our primary goal is to maximize physicians' revenue through several methods, through accurate coding, through efficient management of the overall payment process and our goal is to have a hundred percent clean claims the first time we file a claim in order to get efficient payment and efficient claims handling.
Electronic claims is our best avenue to solve this process and through electronic claims, we are paid at least twice as fast, in most cases three times faster through electronic claims. Our job is made easier and more efficient through standardization of the electronic data handling process. Standardization of documentation or attachments to claims is of great interest to our industry.
The billing industry is a very competitive industry. Over 4,000 billing companies participate in this industry. We need to be efficient in order to remain competitive and even to stay in business, but more importantly, we are being squeezed by doctors to cut our fees to them and so this process would enable us to do that.
The issues you have addressed in your set of questions are of concern to our association in the following areas: The less the attachments, the less the paperwork, the more efficient our process can be. Solving the payment for service, we would address several issues; for instance, fraud and abuse, standardization of coding and we see opportunities for increased standardization in several other areas.
The main areas that we see opportunities for standardization is in your 50 state Medicaid programs. Each of those programs have a different set of requirements that we must deal with. Then your 70 plus Medicare carriers and the areas in which we deal with them and many of them have different requirements that we must deal with.
Then the other requirement areas that we need to see standardization is how we deal with secondary claims and the EOB attachments to claims, which is a considerable problem, because most patients have a secondary insurance and a tertiary insurance. In all cases, we have to file a copy of that EOB with the primary claim.
In many cases, the Medicaid programs require this and they require it in different forms also. The questions that you have posed that we have particular interest in are Question 16. I could go through these individually or we can wait and go through them collectively.
DR. STARFIELD: Why don't you go through them relatively quickly if you can?
MR. SMITH: Okay.
The types of attachments that we currently are required to file with claims are generally pieces of information from the medical record that supports the claim process. We never have the opportunity nor do we have the medical record in our possession and we don't want it in our possession, but we do frequently require information from the provider to support the claims.
Question 17, other documentation that we are often asked to provide are ambulance reports, EOBs for secondary information and so forth that I indicated earlier. The purpose of these attachments in our industry is always to get the claim paid and to justify the claim that is filed.
Item 20, what aspects of these processes would be aided by standardization of electronic interchange? It would greatly speed up the payment process, is the chief issue that we have.
And Item 22, do you ever submit the entire medical record to payers in support of claims? We never do.
Under what circumstances is the entire medical record submitted? I wanted to address that question to show that billing companies do not have in their possession the patient's medical record. When we are asked for further information, we must go through the provider to get that information. He must give that information to us. And this is a rather cumbersome process, slowing down the payment.
Item 24, do you have any suggestions that would assist in the task in standardizing requests for additional information from payers? In the EOB area if, for instance, that we can maintain EOBs on file and just respond to EOBs on an audit basis, similar to the audits we are given by -- must address to Medicare; for instance, once every six weeks or once every two months or once every 90 days that we can make these audits available to Medicaid and whoever requests those EOBs.
And then the interface between Medicare and Medicaid, a considerable number of patients have both Medicare and Medicaid. We must submit to Medicaid a copy of the EOBs patient payment record from Medicare. We feel that we get electronic payments from Medicare, then that same format could be used to transmit data to the Medicaid program. So, if we get a standardization of the payment process between Medicare and Medicaid, then it would be of great help to us.
Item 25, would the automation of health care claim attachment information and standardization of the payers' request for information reduce the operating costs of your facility? As I stated earlier, it would have a significant impact on our cost of doing business, which in the long run is passed to the physician and eventually winds up with the patient's increased cost.
I thank you for the opportunity of appearing before your committee today and I will stand by to answer any further questions. Thank you.
DR. STARFIELD: Do your organizations represent generally the institutional ones or non-institutional or is it mixed?
MR. SMITH: Providers are non-institutional. However, we do have billing companies that do a considerable amount of work for medical schools.
DR. STARFIELD: Okay.
Bob Owens.
MR. OWENS: Thank you, Madame Chairman.
What you are getting passed out to you now is Dr. Rose's statement. If you could read that at your leisure. I would just like to highlight a few things. Again, I do apologize, but this seems to be a very foggy issue and today Chicago was fogged in. So, his plane couldn't land or take off. So, again, I do apologize for him not being here.
If you look at dentistry, the ADA currently represents about 74 percent of all the practicing dentists in the country. That is about 141,000 dentists; so, as far as the marketplace goes relative to dentistry.
If you look at dental electronic claims and the attachments with them, they are not like the insurance claims that you have in medicine. We are not as complicated. Generally, all our procedures encompass everything from the services that were rendered to the supplies and the equipment. So, when we actually submit a service for adjudication or a procedure code for adjudication, all the information that is necessary to adjudicate that claim is pretty much there.
We have far fewer definitive procedures done in dentistry than is done in medicine. What we would hope is that we wouldn't basically repeat the practices that we currently have on paper today, basically look at how we are doing business.
The whole thing that HIPAA is about is administrative simplification and look at what we can do to simplify versus what we can do to basically add more cost burden to all.
The electronic world is basically created to ease the transmission information, to speed up the transmission between the provider and any number of the sources and the payers are really only one of those. You will have prover-to-provider communications. You will have other types of communications that are all necessary to basically happen here in the administrative simplification.
We are looking to streamlining both the diagnostic and the procedure coding systems by some recent things that we have done in basically creating the SNOMED Category System. We are dramatically looking at cutting down on treatment costs just from the aspects of being able to basically manage the information that we do actively get on the claims itself.
There are approximately 333 million dental claims filed annually. They cost currently about $1.50 in staff time. When you do that stuff electronically or these claims electronically, the cost drops to about 30 cents in staff time. So, it is a pretty dramatic effect as far as the overall cost of administration. And we would look at anything to basically counteract that or add more costs back into the system as more of a negative than a true positive because in most of this, it is simply because the information is going between one computer into another.
It can be conclusively demonstrated that attachments increases both costs of the claim not only to the provider but to the payer. On the provider's side, office staff have to assemble the information. They have to codify it. They have to either print it or hard copy and mail it.
On the electronic side, the computer equipment, most of the dentists possess, basically cannot do radiographs or that type of equipment. It is important to note that probably less than a third of dentists actually have computer systems.
So, when you say you need attachments and you are looking to make those national standard requirements kind of thing, that is a very big learning curve for dentistry. The technology to actually do that even if you impose such a standard is moderately to very expensive. There is a lot of -- dentistry is made up typically of one practitioner offices. So, you know, any cost burden that you put on them, these are all small businessmen that are required -- you know, that are running at very cost effective operations and to put undue cost just for the sake of being able to collect attachments with no real value is problematic.
It requires a significant amount of ongoing information, technology support. It is not a matter of just simply dumping the systems in and walking away. Someone has to physically support that. If you are going to be reading images and free form text and binary files and everything, you are going to have to have some sophisticated software to do that and the storage capability.
When you get into x-rays, yes, even though your teeth are smaller than your chest x-rays, over time, you are going to fill up a database and you are going to fill up some memory. So, again, the costs just continue to escalate into real value.
Software, at the present time, the provider is much further ahead, though, in digitizing its information because the third that do have computer systems, in dentistry they are more clinically focused than they are business focused. So, they do have a lot of imagining and a lot of clinical applications inherent as part of the actual practice itself.
Software, if you look at the capabilities, until attachments basically become less expensive for the practitioner and the payers to adopt, hard copies attachments, you know, we are going to continually have that nightmare. The logical question is how big of a problem is the lack of attachments on an electronically transmitted claim.
The answer, I would submit, is probably not much. Since we have gone to electronic claims processing, payer requests have greatly reduced. And in some cases, the payers have just said because we are doing communications, you know, basically computer to computer, the error rates are far down.
So, we basically don't need attachments anymore. Keep your x-rays. So, we are looking to try to basically eliminate the human error and the human intervention and maximize the efficiencies that we can gain from a true administrative simplification, electronic process.
It would be a mistake to require routine submission of any claims add-on material because basically, again, all you are doing is escalating the cost to the patient, the daily treatment and delaying the ultimate goal of administrative simplification.
Thank you for inviting us to testify in front of you today and, again, my sincere apologies for our president-elect.
DR. STARFIELD: Thanks very much.
Jerry Henry.
MR. HENRY: Good afternoon. Thank you for allowing me to appear before you.
I am Jerry Henry, chairman of the Kentucky State Uniform Billing Committee. I have served on this committee since its inception in 1982 and have been chairman since 1984. I am employed by Baptist Healthcare System based in Louisville, Kentucky.
Our system includes five hospitals located in Louisville, Lexington, Paducah, Corbin and Lagrange. These facilities range for 130 beds to more than 400 beds with a total of more than 1,500 beds for the system.
Two of our facilities have Medicare designated separate psychiatric and rehabilitation units. All five facilities have skilled nursing facility units. Three of our facilities routinely treat patients living in Indiana, Tennessee, Illinois and Missouri, as well as Kentucky.
I am directly involved with our business offices in the computer support functions for both paper and electronic billing for all five facilities. Our system was a pioneer in Kentucky in developing electronic claims with Blue Cross of Kentucky in the 1970s. We have been filing commercial electronic claims through NEIC since 1980. Today, we file electronic claims to any payer who is able to accept them. Out volume of electronic claims would be significant increased if we could eliminate paper attachments.
Kentucky passed Uniform Health Insurance Claim Forms legislation in 1982. At that time the kentucky Hospital Association asked the insurance commissioner to delay implementation of the legislation until the UB-82 claim form could be started. The commissioner agreed and issued an order to that effect on July 1, 1982.
On July 8, 1993, the commissioner issued a new order that dealt with the implementation of the UB-92 claim form. The state manual details the attachments that are still necessary. Providers are asked to notify the Kentucky Hospital Association if a commercial payer attempts to require additional attachments or certifications.
Kentucky is a no-fault auto insurance state. The rules used by Medicare in the Medicare secondary payer arena have been adopted by every payer in the state. Every payer requires a copy of the auto insurance EOB or denial before processing the claim.
Our Kentucky rules work well except when dealing with self-insured plans or others hiding behind ERISA. There are also problems with out-of-state payers who are not licensed in Kentucky. While we have a uniform claim form, we do not have uniform billing.
Federal plans, such as Black Lung, Champus, CHAMPVA and Federal Workers Compensation are among the worst offenders in requiring attachments or requiring hospital providers to file HCFA 1500 forms rather than accepting UB-92 claim forms. These plans also require a variety of attachments. Likewise, these payers have been slow to accept electronic billing.
State Medicaid agencies cite federal law as requiring Sterilization Consent Forms, Hysterectomy Consent Forms and copies of EOBs for any prior payments on a Medicaid claim. Kentucky Medicaid requires a copy of the spend down letter, which they issued, be attached to any spend down claim. Their system should be revised to hold this information rather than require an attachment.
Many payers apparently require attachments to cover deficiencies in their internal systems. For example, Anthem, formerly Blue Cross of Kentucky, converts both paper and electronic UB-92 formats back to UB-82 for processing. There are many elements unique to the UB-92 that may not convert. This may also explain why some commercial payers require their own form attached to the claim.
Federal legislation is needed to control attachments and the effect of attachments on payments. We have accomplished nothing if we forbid attachments and payers hold claims waiting for the patient or subscriber to complete and return their referral form. We know of some payers today who are guilty of this.
EOBs should not be required when payers begin forwarding adjudication information to subsequent payers. The rules need to clearly state that subsequent payers must accept the electronic format and not require separate claim submission with copies of EOBs attached.
An excellent candidate for electronic attachment is the ER report that New York State has added within their system. The penalties in the HIPAA legislation need to be revised. The current monetary penalty will not encourage payers to revise and update their systems. Creating standard electronic attachments will have little effect unless payers accept them.
Attached to my statement are the specific answers to your questions for presenters. And, again, I thank you for this opportunity to share this information.
DR. STARFIELD: Thank you.
Jean Narcisi.
MS. NARCISI: Thank you.
My name is Jean Narcisi. I am the director of the Office of Electronic Medical Systems for the American Medical Association. It is my pleasure to appear this afternoon on behalf of the AMA, before the subcommittee. I would like to thank you for the opportunity to testify.
My statement summarizes the views and concerns of the AMA and its members regarding the issue of standardization of health care claims attachments addressed by the Health Insurance Portability and Accountability Act of 1996.
The following comments will address many of the provider-related questions we were asked to discuss, as well as any other issues that could be of concern for the medical profession.
In response to the question regarding the definition of a "claim" versus an "attachment," the AMA definition of a "claim" is the submission of information by a provider or covered individual to a third-party payer using a standardized format, either paper or electronic version, sufficient to establish that covered health care services were provided.
The claim includes a request for payment or reimbursement to the provider or covered individual. An attachment would include information either requested by the payer based on a prepayment or postpayment follow-up or provided by the provider at the time of submission of the claim. Attachments consist of information, presumably not available on the initial claim or claim format that provides further supporting details of the claim.
With respect to the question regarding differentiating the information appropriate for the claim versus the information appropriate for the claims attachment, the AMA believes that a claim should be based on a straightforward minimum/maximum set of well-defined data elements and should be sufficient for payment in the vast majority of submissions. The claim should also allow for a limited set of additional data elements to be included, such as additional information to specify any extenuating circumstances or justifications that may assist in the review of the claim.
We believe attachment information should accommodate a variety of paper and electronic technologies and should allow for structured data elements, images, and free form text if necessary. In addition, attachment information should only be requested for specific types of services and the requests should be limited to data that cannot be accommodated in standardized electronic and paper formats.
Furthermore, the AMA believes that providers should have the ability to submit attachments at the time of claims submission and should not have to wait for a payer query or specific attachment requirement to explain, for example, the unusual circumstances associated with the pattern of treatment.
In response to the question regarding the types and formats of claims attachments currently provided to payers, AMA member physicians are currently asked for a variety of supporting information to adjudicate a claim. This can include information that is contained in the patient medical record, such as operative notes, test results, patient consent forms, the medical necessity of ordered durable and medical equipment and the cost and duration of the use of drugs and supplies, et cetera.
Requests for administrative information, such as copies of a driver's license, have also been requested. Submissions can sometimes be made electronically. However, attachment information is sent in manually on paper. In some instances, payer requests can be satisfied via the telephone or fax.
In response to the question regarding the purposes of these attachments and the other documentation, we believe that the payers may need additional information to assess coverage or medical necessity for specific types of services or to meet specific contractual or regulatory requirements.
All too often, unfortunately, our members have concluded that these requests are intended to delay payment of claims or to provide a basis of unwarranted denial. They may also often reflect a desire to pressure physicians into billing at a lower level of service.
Requests vary considerably across payers, causing delays and significant costs for our members. However, the AMA believes that considerable strides can be made in the attachment issue if the following are taken into consideration: Standardization must apply to all public and private payers and cover both paper and electronic formats. It is essential to standardize when attachments are to be required and not just how they are to be submitted. The AMA is very concerned that lack of standardization in the situations when attachments will be required, coupled with standardized electronic request formats, will lead to an increase on the burden placed on physicians and on the threats to patient confidentiality.
Also, limit the number to a well-defined set of possible attachments.
In response to the question regarding the process that would be aided by standardization and electronic exchange of information, the AMA understands that standardization of formats and electronic exchange should reduce the cost of preparing and submitting attachments and enable physicians and the health care system to realize the full benefits of electronic data interchange and administrative simplification. At the same time accomplishing such partial standardization without standardizing when attachments are required could make matters far worse.
The AMA supports the position and the efforts of the National Uniform Claim Committee and the National Uniform Billing Committee to use the claim format to eliminate the need for specific types of attachments through efficient use of specific data elements. With this vision, there would be substantial cost savings for providers and payers and a realization of the true potential for administrative simplification.
With respect to the question regarding the relationship between claims attachments and the medical record, quite often, attachment information is drawn from the medical record. However, we believe that attachments and their standardization should be viewed primarily as part of the administrative and claims-related process.
Physician practices often submit the entire or large portions of a medical record to support a claim. The decision on what to submit will depend on the extent of the payer's request, the physician's expectation of documentation that will be needed and the extent to which the information in the record is in the format requested by the payer.
The AMA would like to suggest the following regarding standardizing the request for additional information from payers. First, the instances in which payers can require attachments must be standardized with the goal to reduce the need in the use of attachments.
Second, standardized mechanisms to submit attachments should allow for use of both free form text and standardized messages and codes for results, et cetera. And the standards should accommodate the range of current administrative and clinical information systems in a physician's office.
In response to the question regarding reducing operating costs for providers, the AMA believes that there is a substantial opportunity for cost savings if the attachments are standardized along the lines that I have just suggested and if this standardization is not tied to particular technologies or levels of practice automation.
We cannot stress enough, however, the potential for systems upgrade and training costs, coupled with increased practice costs if payers generate an increased number of automated attachment requests that require non-automated or semi-automated responses.
A recent study commissioned by the AMA determined that, in general, the use of electronic medical records systems in the health care arena is relative to the size of the practice. The larger the practice, the more likely the use of an electronic medical records system. However, it was found that with the ambulatory practice, less than 5 percent of outpatient records were computerized as of 1997. Group practices of 20 or more had 20 percent of their patients records automated and 25 percent of the hospitals had their records automated.
In the same study, it was determined that the electronic medical records systems ranged significantly in the cost from $4,400 to $775,000, depending on the size of the installation. In addition to these purchase expenditures, the physician should also expect an initial increase in administrative costs, as well as a decrease in productivity while physicians and staff adjust to the newly installed systems. These are very important factors that need to be taken into consideration.
In response to the question regarding who should have the responsibility for approving the content of the claims attachments, we believe that there should be an organization, other than those responsible for the messaging transactions, responsible for the messaging transactions, responsible for approving the content of claims attachment transactions.
This organization should be representative of all parties affected by health care electronic data interchange, providers, payers, standards developing organizations, regulatory agencies, et cetera. Based on their structure and the current and anticipated responsibilities, the NUCC and the NUBC are appropriate to assume this task.
The NUCC has an official operating protocol that provides full due process, open meetings and the ability of non-members to generate agenda items. The protocol is currently being revised to broaden the parties represented on the NUCC and to address the needs for an additional appeals process.
Data content maintenance should be conducted through the kind of public/private partnership that these two committees exemplify and not viewed as a primarily technical communication standard.
In response to the question related to the use of HL7 messaging in the standardization of attachments, we realize that HL7 is widely used in current health care software systems, especially in the hospital environment. In the AMA commissioned study that I referenced earlier, it was found that 46 percent of health care software systems support HL7. However, use of HL7 by physician systems may not be sufficient at this time to base an attachment approach on this messaging standard.
In addition, it is not clear to us that the query response approach to the attachments being developed by the claims attachment Proof of Concept Project, to be discussed later today, will fully or adequately meet the needs of practicing physicians.
Thank you for this opportunity to present the views of the AMA. I would be pleased to respond to any questions that you may have.
DR. LUMPKIN: Can we get a copy of that study?
MS. NARCISI: Sure.
DR. STARFIELD: Okay. The floor is open for questions, clarifications, discussions, position statements.
DR. MC DONALD: Bill, what is the date that attachments would actually come to be effective? That is, how much are we projecting into the future in terms of usage rates and the rest?
DR. BRAITHWAITE: The HIPAA law requires us to adopt a standard in February of next year and then the participants in the health care system have to comply with that standard two years thereafter.
DR. MC DONALD: So, that is 2001.
DR. BRAITHWAITE: So, February of 2001.
DR. MC DONALD: Three years away?
DR. BRAITHWAITE: Yes.
DR. STARFIELD: I guess all the hearsay was true, right?
DR. MC DONALD: I thought these were very insightful set of presentations and a lot of facts and a lot of experiential things. And it sounded like -- I guess to summarize the statements was, you know, ban attachments would be sort of an easy way to say it in one lump.
Some of the arguments were, to me, compelling, although I am an academic physician, so I don't really have to fuss with it. So, it is not really a personal thing to me. But it sounded -- some of them were reasonable and set some rules about whens, rather than whats. I mean, some of the stories about how the payers are asking for data that they have, you know, seems like there is some early, you know, sort of low hanging fruit, you know, if you will.
But what mechanisms are there really that we could -- as a committee could help encourage or even in terms of the HIPAA legislation, does it speak to these things in a way that we could put a wedge in or help make some of these stuff work better?
MR. HENRY: The UB-92 format allows form locators to report prior payments by other payers and, yet, my hospitals tell me that in most cases, they don't even re- bill the claim. They just pick a copy of an old claim and attach the EOB to it because their experience is that without the EOB, it is not going to get paid and even if they put it on the claim form, the payers don't pay any attention to it.
Now, that is different than the 1500, which doesn't have clear blocks for prior payments for each payer.
DR. MC DONALD: Yes. I understand that whole business about the prior payments is under discussion in the industry and there are some proposals to not think about those. I may be speaking out of school, but not think about them as attachments because they certainly -- there are better ways to deal with that.
Maybe some other folks could speak to that, but I thought there was some movement in the industry to deal with the coordination of payments and benefits and all the rest of that.
DR. CHENG: Coordination of benefits has always been the biggest game in town, that that is out there, as far as I am concerned, to confuse providers. One piece of legislation in one state says you do it this way. Patients cross state lines and now all of the sudden you have to do it another way and the Feds say do it this way and someone else out there says do it that way.
It has reached the point where I don't care. I don't care how -- really. I really don't care. You want the information. I want to get paid. Okay. I have got payrolls to meet. I have got vendors I have to pay. I have to keep the hospital open one way or the other. You want it. You got it. Okay?
And I don't care what it is. If you have sent -- like I said earlier, if you have requested information once before, I am going to flag it in my computer system. This sucker wants everything. Send them everything, whether they need it or not. They are going to get it. If they can take it electronically, great. If they can't, hit them with the paper. Bury them.
And, hopefully, they will come back and say wait a minute, we don't need this. Okay. Great. But it is surprising, too, this morning, listening to the payers, what was very interesting listening to the payers was they all have contracts with Medicare.
I know what Medicare wants. I know what Medicaid wants. Tell me about your private business. What do you want there? How come there wasn't a real managed care company saying, well, we do this because we want this and that and this and that. Okay. Tell me, you know.
Right now, I am speculating. I would say, looking at the composition of the panel this afternoon, providers have more of an interest than the payers. I wonder why. And being a person of few words and not opinionated in any sense -- you know, I come from a small hospital and I will do everything and anything that is necessary to speed up the process for both institutional claims and the physicians' claims. I have got well over a hundred physicians on salary on staff. And we submit claims for them all the time.
I keep hearing the complaints from them over and over and over again, why do they want this? And I just tell them, I don't know, other than they don't want to pay it and protect their profits. It sounds plausible, since I don't get any explanation.
I have got limited electronic medical records, but what I am able to do -- I have got scanners all over the place. So, I will scan the charts. Drop it down to an ASCII format. You as a payer, if you can take it, great. If you can't, I will send it to you on a diskette.
If you don't want to do that, all right, fine. I will give it to you on paper. I have got templates built into my computer system of attachments. So, the end user, my billers, my medical records staff, my admissions staff, they all report to me. Fill out the information just in case it is needed and if it is one of the payers that is on the list, my infamous list, give it to them. If they are not on the list, let's wait and see and if they make the list, fine. Hopefully, they don't. But whatever they need, they will get.
What I would like to see all of you do is come up with standards. Don't allow any exceptions. I have gone through this for 25 years. I am getting tired. I am not young anymore. I would like to see standards sometime within my lifetime to make it easier for the providers. I don't want to hear about fraud and abuse. Most of us, virtually all of us out there are trying to do the right thing. It is only a handful of idiots that need to be put behind the bars.
Once we get rid of them, then we have got some trust between us and we can do things and do it once and get it right the first time. But when you come up with the standards, make them stick. And I would say if you can, say these are the maximum standards, not minimum, because the moment you say "minimum," anyone can say, well, you got the minimum, I need this, too. Say "maximum." Then they can't add anything else to it.
If you are going to make change, please don't make them retro. You want to see a provider go nuts, make a change, make it retro and watch us do all the song and dances and everything and jump through the hoops. But do something, please.
DR. STARFIELD: Jean.
MS. NARCISI: I think one of the things that this committee could do is work with the content committees and really -- on defining what an attachment is. That is one of the things that was really hard to pull together when formulating the responses to these questions, well, what is an attachment.
The other thing is is that the way that, for instance, the 837 transaction has been designed, it includes claim information, but it also includes a lot of extra stuff that is probably considered an attachment, but at least it is there and it is probably workable in an electronic environment, but you need committees, like the content committees, to actually pull together some type of an implementation guide on, for instance, defining these are the data elements that are needed for an ambulance form and those kinds of things.
So, I think by using what is already there and what works, but then moving forward with a model that will clinical information to be captured, but I don't think you should use, for instance, the Proof of Concept Project that is being developed right now, use that as developing a regulation. I think that is a very necessary thing that needs to be done, but I think we need to wait until the results are finalized before we turn it into law.
DR. STARFIELD: Dan, I wonder if you would comment on that. You testified that you try to anticipate what the payers want and you almost never submit medical records. How do you figure out what it is they are going to want?
MR. SMITH: Billing companies, as such, do not submit medical records. However, the companies that do practice management for physicians have under their control the physicians' medical records. So, that is the -- when you ask what specialties we bill for, we bill for all specialties.
Our company specializes in the hospital-based physicians, radiology, pathology, anesthesia, emergency room and so forth, but other billing companies do all specialties.
I would like to speak just a bit about Paul's comments. Hospitals have a lot of money.
DR. CHENG: No, we don't.
MR. SMITH: The way we view them they do and they have a lot of expertise in dealing with electronic claims submission and so forth. I talk with the Medicare carriers. For instance, the Medicare carrier in Richmond tells me that billing companies should be getting rich. We should be doing more new customers than we can handle because there is no doctors' offices out there anyway that can understand what they are doing and that can manage their computer system.
He tells me that about 90 percent of the offices have very little, very limited capability in their computer system. They have Jane, who they sent to the school when they bought the computer system, and they gave her a manual that thick and Jane brought the manual back and put it on the shelf and hasn't looked at it since.
That is your capability in most doctors' offices, as I see it. Anything we do in this area of electronic data interchange has to address the small office capability. I will give you one example of how this can -- this transpires. In our area in southwest Virginia is rather rural. An enterprising physician formed a group and he called it the Blue Ridge Health Network and he brought in 80 primary care physicians. And under that umbrella of 80 primary care physicians, they were able to buy a very sophisticated computer system and put each one of those offices on electronic claims and bring them up to date and now they are even working on electronic medical records.
So, as we see more and more consolidation in this area of the industry, we are seeing greater capability, but I will assure you the way it is right now, about 30 percent of the doctors' offices have that capability. Seventy percent of them have very limited capability of dealing with that type of information.
So, anything that we do, we should address it in that -- take care of doctors.
DR. COHN: I was actually just going to ask a question to Mr. Henry, only since there has been discussion throughout this panel about the issue of controlling the amount of attachments, putting some rational framework into all of this. And I sense that despite your complaints about groups of insurance companies that didn't abide by your regulations that, at least within Kentucky, it sounded like you had some control over what was going on.
Can you talk a little more about how you approach this within the areas you do have control within Kentucky and your satisfaction, dissatisfaction with what is going on?
MR. HENRY: Well, as the UB-82 came into being and our committee was comprised of about 25 people across the state representing providers, payers, Medical Records Association and various other -- the insurance commission was represented, and what we tried to do was determine what attachments were really needed.
Now, at that time, Blue Cross of Kentucky had a COB form that had to be filled out and attached to every claim form. A couple of us started looking at it and we said to the Blue Cross rep, you don't need your COB form and we can show you why. It is all already on the claim form.
Well, he looked at the form. They agreed. Blue Cross did away with their COB form at that time and we were able -- once we got that one started, we were able to eliminate everything except what Medicaid was requiring under the federal guidelines; hysterectomy, sterilization, EOBs, spend down letter and so on.
Now, that only covered those payers who are under the -- by law are covered by the insurance commissioner; that is, people to whom he regulates and issues a license to. It did not cover the self-assureds, who were not very many at that time. We are talking now 1982.
Since then, the self-insured plans have increased dramatically and they hide behind ERISA. They are not subject to the state laws. So, they can do anything they want to and they do and many of them came into being not even understanding what they were proceeding to legislate and they were asking for attachments that made no sense. They still do.
There are cases today where our business offices -- we are just like Paul's group. We know who they are and we routinely attach the information up front because we want the claim to be paid.
DR. MC DONALD: I have learned a lot. I have been working mostly with the payers and they are a really good lot. I didn't know that before.
Both sides, though, you know, kind of want the other side to clean up their act, you know. So that it is kind of -- that is how life is, I guess. But the thing that came out that I didn't realize is not the ERISA thing. If you have a gas station, you can self-insure and then you become your own insurance company.
So, I guess there is a systems first, trying to think about what we can really ever do with this. I go on the one hand, the insurance industry won't do it anyway and they don't have to because millions of them don't have any regulatory -- you don't have any clout on them. And the physicians don't have enough computers yet and we are projecting into the future. So, they are not going to have the electronics thing.
I mean, what are we doing this for? We need to come back, I think, and say -- we need some rationality. I think the only way you can do this is we catalog the questions. We get rid of the 5,000 variants. It is going to make life easier for everybody, but that means that somebody can't do something they could do before and they will push back.
So, what does the legislation say? Can we push -- can we do anything that is going to not get pushed off the table by them pushing back?
Does ERISA get free of all these regs?
DR. BRAITHWAITE: No. The regulations that we will put out about claims attachment applies to every health care plan, including all ERISA plans. So, they do not get out from underneath this. They must comply with the standards that we set.
DR. MC DONALD: In all the discussions we had, you know, on the insurance end, this is -- these are sort of the good payers, I think. I mean, the bad ones don't come or something. I don't know. Because once you meet someone, they are always good.
But they say that, well, yes, we could standardize but they won't accept that. I mean, there is always -- these other guys won't accept that. So, I am just really -- and then when you look at the questions, it is a two year's job if you were working full time at it with 20 people, see, because there are forms out there that haven't even surfaced yet. The Medicaid, apparently, are the terrors of the deep.
[Laughter.]
We kind of enumerate the ones that are kind of out there and they always say and then there is Medicare and everybody kind of blanches, you know, and gets --
[Laughter.]
MR. MOORE: That is run by the state.
[Laughter.]
We are just custodians of their problem at HCFA.
DR. MC DONALD: If one specifies limits on this or that, can it apply to them, too?
MR. MOORE: I think what we are wrestling with here once again is trying to define what the problem is and I don't think that we are at a point in the regulations process that we have done all that. Part of what we are hearing, there is staff out in the audience who are working -- who are going to be working on those regs, taking all of this in as we proceed on that.
So, it is not -- this part of it is not over and you get to help define that. Where we are going with some of those attachments, I think, we have heard before that, you know, we need to proceed, identify what is important, try to start, you know, wrestling with those pieces and not addressing some of the other things that are going to be further down the road.
DR. STARFIELD: Bob, did you have an answer to a question?
MR. OWENS: Well, I just was commenting to Clem, basically, there is like four pieces. I think the first piece is you are absolutely right. I think the goal that we need to have and we would hope the committee would have is to basically eliminate attachments, specifically, and demonstrate they are not necessary.
Beyond that, if you need to deal with the here and now and how do you get there, there are really three pieces or three steps. I think the first step that we have to look at is defining what is the business process or the business scenario in which you need an attachment.
Once you define that, then you can go into what is the content that is absolutely necessary and only absolutely necessary to accomplish that business process. Then, third, actually develop the standard or look to adopt a standard that is in place that meets that content criteria.
But I think where we kind of jump right into it is we want to jump right into defining "content." We want to jump right into defining the standard without dealing with the fundamental issue and that is standardizing the business process.
MR. MOORE: I think we have to be careful here when we start saying what is the business process. It is not just what the payers pay the provider. I know at HCFA, I move from one area to another, where we are looking at the clinical standards and we are looking at the quality of care. That has not occurred to the -- we are way behind our curve on that and doing a good job in that area.
I don't think we have begun to wrestle with the issues, the data that is required, in order to make that assessment and as we -- I made a statement before you came, Bob, about we are having a problem in making policy on what is a true EPO for dialysis patients.
It is not necessary to have that at the time you make payment possibly, but you -- where do you get this data? How do you organize it? And if it is not in the claim and if it is not in an attachment, you have to have that -- you have to start making decisions based upon evidence, not subjective, making decisions as to what is good care and where we are going with policy decisions.
So, somewhere we are going to start pushing that and we are probably going to be looking at that when you -- when this committee looks at the clinical record. But then it becomes a question of what is in the clinical record and how do you use it and we can't wait until the year 2010 for Congress to adopt whatever this committee is going to recommend.
DR. MC DONALD: There is another piece to it and that is there is clearly -- I mean, just by looking at operational reality, some of these things, everybody uses them and like there are these ambulance attachments. But the other challenge is figuring out if one agrees one needs it, how do you say the questions? What do you really mean by them? This can be done independently -- I mean, maybe in parallel with some of the other analyses, but this is tough. I read some of these forms and it is hospital, one, star -- two star -- you go back and forth. You can't tell really what happens if they have passed, if they are going back and -- I mean, you just can't tell what you are supposed to say there.
Now, the HCFA ones are actually better than most, I have to say.
MR. MOORE: Thank you.
DR. MC DONALD: Once a year we try to get the balance even, you know, on this thing. But I think we need -- I mean, in some of these -- and the other part is is that if the payers or even the providers in abstracts say what we want to know is sort of a conclusion, well, as we project into the ten years from now, a lot of that stuff is going to be available as raw data and it will be easier for everyone to get it that way.
If we don't keep aware of the difference between conclusions and raw data, you know, some of those things that used to ask for hypercholesteremia, they can't look at their old data anymore because they don't -- it is a different threshold to what -- I think at the start of the cholesterol, they would have been -- it would have been easier to collect and better off.
So, there is a complicated mixture that we have got to stir to get this to work right. But it is encouraging, though, that -- but right away, we should say, you know, we are going to not -- I think we have got to say "not," too, like -- I mean, we have got to start saying this is what we will do; this is what we won't do. And this business about everybody can -- literally, some of the payers say we make up a new rule next week because we detected some possible fraud thing and we ask for new kinds of data.
But the other side of the world can't live in that kind of a pot. Maybe everybody doesn't agree that they will give up -- both sides will give up something and we could get something easier for everybody.
DR. CHENG: I will give up whatever you need. Just pay it. Pay the claim.
DR. MC DONALD: I was worried about Paul. Paul was at this meeting in December and when he gave this beautiful, smooth presentation, I thought he had a personality transplant.
[Laughter.]
Then he started just kind of kicking around here. I thought, oh, he didn't change. He is all right.
DR. CHENG: You know, 25 years of dealing with attachments and the claims process, I look at these attachments. The questions are the same. They are not different. They are essentially the same, maybe an additional comma here, putting in "the" or "a" or "please list" --
DR. STARFIELD: What are they? What are they? Give us an example. What are the common ones?
DR. MC DONALD: I can give you some.
DR. STARFIELD: No, let him.
DR. CHENG: How long was the duration of the treatment? Justify the treatment. Who are the participants of the plan of care? What is the sex of the patient? What is the age of the patient? Are there family members involved? Are there -- is there information -- when did the problem or crisis occur? What are -- I am usually not at a loss for words.
Dates of service. Type of service. What is the level of service? When was the patient last seen? If it is a female, last LMP, regardless of age. For a senior citizen, I have seen some ridiculous questions. A 98 year old female that had a total hysterectomy, did you advise the patient that they will no longer have children?
MR. MOORE: That was one of our finest carriers.
DR. CHENG: At the hospital, we have seven ambulances. Okay? And the ambulance requirements are totally absurd because, yes, we have seven ambulances that roam the city of New York, along with a couple of hundred others under the control of EMS, emergency medical services of the New York City Fire Department.
So, for example -- well, even now with the Medicare requirements, give us the mileage from point A to point B. These are roving ambulances. We have absolutely no control over them, even though they belong to us. So, everyone, whenever we send an ambulance claim in will ask the same questions, mileage and all the rest, and we tell them we are under 911. We have no control. Tough nuggies. You are not going to get paid.
MR. SMITH: Why didn't you take them to the nearest hospital?
DR. CHENG: Yes. Why didn't we take them to the nearest hospital? They did go to the nearest hospital. We would like them to come to our hospital, but the fire department says, you know, because you are patrolling this area or cruising this area, you go to that specific hospital at that time if an emergency occurs.
If it is a non-emergency, that is fine, too. You know, you have to go there. The Point of Concept Team has already started gathering all sorts of copies of various types of attachments and sundries and if you take a look at the questions and just ranking and looking for the similarities, you will find that they are almost all the same, just slightly modified.
It is like the cost report. The cost report in one area is one way and then you go somewhere else, it is a different way and a slightly different format and all the rest. As much as Medicare has done -- has made great strides as far as standardization is concerned, there is still not one real standard.
MS. LAUER: I just had a question. I am not quite sure how much has been done up until this point, but is it reasonable for prior history information to expect the payer populations to pass information from one payer to another as a patient transfers from one plan to another to be supported or do you want hospitals and providers to be the ones that get them all the records all over again?
I mean, I think of it from the perspective that if a patient -- if I as a patient move from Plan A to Plan B, I would like Plan A not to have my records anymore and have my records with Plan B, not Plan A and Plan B. I mean, employers change plans annually if not every six months, depending on the best rates that are going around. So, you could theoretically have quite a number of insurance carriers having your medical records everywhere.
If a lot of this clinical information, which I heard this morning, was a lot of clinical data to support, could it not be -- let's take a very extreme example, like a liver transplant, where someone has been sick for years, up until they have become, you know, ready for a transplant. That information resides somewhere already in somebody's system, probably their own, but, I mean, you know, again, that would justify the services.
Is it not reasonable to think about standardizing transfer of information within an industry itself?
DR. LUMPKIN: I suspect Bob Gellman's committee's answer would be "no," in that people who are concerned about privacy tend to want to make the decision about sharing of records. So, that becomes a little bit more complex. Just because I change from plan to plan, I may not want them to have records about some treatment, which isn't pertinent to that plan.
I have a question -- Clem, did you have a comment?
DR. MC DONALD: It was a follow-up on that. Actually, if that was a position, I think that would be tremendous. I think we ought to do it. But it can't happen until the next phase of HIPAA. So, we have got to wait and see how that plays out.
DR. STARFIELD: Why don't you clarify what that next phase is?
DR. MC DONALD: Well, there is an electronic medical record phase we are supposed to look at. I mean, I think certainly one -- a patient should have a chance when they signed up with the alternative company to control it. So, I don't think it really conflicts with the privacy. But it is absolutely ridiculous especially with the churning that goes on now. The physician comes in and he is totally blind. So, he has got to order a bunch of stuff or doesn't because he doesn't want to charge -- you know, spend the money. So, either way, it comes out wrong.
DR. CHENG: That is going to take time, the electronic medical record. We are getting set to install an electronic medical record system at our facility. The cost is incredible. It is absolutely astronomical.
DR. STARFIELD: Back to you, John.
DR. LUMPKIN: With the prior panel before the last panel, I asked the question of what percentage of the claims result in an attachment being requested or additional information. The number they gave us was 3 and 20 percent. How does that fit with your experience?
MR. SMITH: I would like to say it depends on the area you are operating in. It can get 3 percent up to 20 percent, but over the years -- I will give you one example of a payer. United Mine Workers Health Plan, in the sixties, if we could get a piece of paper to them with any kind of information on it about a procedure and a cost, they would pay it.
As their money ran out and it became more difficult for them to pay claims, they started generating more and more bureaucracy, more and more requirements to get claims paid, until today they -- for instance, to get a radiology claim paid, we must have on file with them the radiologist's name and he must be board certified and he must have attended a qualified medical school in order to get his claim paid.
What I see as you see that escalation of requirements, what this additional information required in attachments is just a way to beat down the community that files claims to reduce the amount paid to them as much as they can and discourage them as much as they can from filing the claims at the level they should be paid.
So, it is a game. It is an economic game between those who are filing claims and those who are paying the claims. The payer wants to pay as little as possible. The provider wants to be paid as much as possible. That game will continue as long as we have the escalation of medical care that we have in the United States.
MS. LAUER: In our organization, I would say it is probably closer to around 15 percent. However, keep in mind that that 15 percent may come around -- equate to about 75 percent of our overall revenue, which is being held for those claims attachments. Okay? So, your inpatient, your emergency rooms, your ambulatory surgeries are the ones that are being -- all the records are being asked for.
So, it is the majority of your revenue that is being tied up.
MR. OWENS: In dentistry, the low end is zero percent. There are a couple of payers that are actually doing tests of no attachments; basically, just adjudicating the claim based on submission.
MR. HENRY: In Kentucky, our Medicare intermediary crosses over the claims to Blue Cross of Kentucky, which is now Anthem, and also to Medicaid. That means that most of our problems come with the other payers and where someone else is secondary to Medicare, other than Anthem or Medicaid. I would say probably 20 to 25 percent of our claims require attachments.
DR. LUMPKIN: And is it the same, representing 75 percent or a large chunk of your revenue?
MR. HENRY: No, probably not because most of our hospitals are rural or almost rural and 60 percent of our business is probably Medicare.
MS. NARCISI: The AMA hasn't collected that type of information, but the NUCC did pull together a survey on attachments. I don't think the data was ever actually finalized, but one of the chairs of the Attachment Subcommittee said that it was probably over 20 percent that he saw in the results.
DR. CHENG: One observation that I have been finding, as Medicare managed care expands and Medicaid managed care expands, each of these managed care type companies that handle Medicare and Medicaid are asking for more and more and more attachments. At least, with Medicare and Medicaid under the traditional programs, we knew up front. We know what they need and we will do it. Now, all of the sudden with managed care taking over Medicare and Medicaid as separate programs, they are just coming right out of left field asking for more and more information. It is amazing.
DR. LUMPKIN: Maybe as a follow-up, to what extent are you seeing things like HEDIS forms and HEDIS data collection being put in the guise of attachments in other kinds of quality assessments.
DR. CHENG: Quality stuff?
DR. LUMPKIN: Yes.
DR. CHENG: You know, in my other life, I am the adjunct faculty staff at St. John's University and Cedarview Post Long Island University and I teach in the health care program. And I always am asked by my students about what is quality and I give them an example. If I am your physician and I hold your hand all the time and I talk to you all the time, you are not going to care about whether I give you really good care because you really don't know.
And in your mind, that is good quality care, but in everyone else's mind, it could be the worst thing in the world. Now, when we start getting questionnaires from all sorts of companies everywhere, and you can tell that they are HEDIS related and all the rest, and I say to the medical records staff and the physicians that ask me don't worry about it. Just fill it out. Fill it to the best you can. Get rid of it and go ahead and take care of the patients because that is what you are there for. And don't worry about the quality of the information that is going in.
DR. STARFIELD: Kepa.
MR. ZUBELDIA: I have a question on the percentage of claims that need attachments. If I am hearing correctly, you are talking about maybe under 20 percent and, however, I understand that between 30 and 40 percent of the claims have a secondary payer. You have testified that those secondary payers will not take the secondary claim without an EOB attached.
Are you counting in your percentages those EOBs as attachments or are you not counting the EOBs as attachments or maybe are you thinking that the secondary claims are the responsibility of the patient and they don't count at all?
MS. LAUER: I am not counting EOBs as an attachment in the numbers that I gave you. Those are the primary payers that are asking for those initially. It would be another payer then behind that. So, no, I did not include EOBs, per se.
MR. ZUBELDIA: How about the other members?
MR. CHENG: I didn't include EOBs in mind either. But if you want to include EOBs, then it is like 90 percent of the claims.
MR. ZUBELDIA: So, if you include EOBs, what percentage would it be for the other members of the panel?
MR. SMITH: In our area, it would probably go up to 40 percent in some areas.
MR. ZUBELDIA: It sounds like very ripe, low hanging fruit.
MR. CHENG: It is also depending on the service. I mean, my facility is an HIV-designated facility. And all the services that we render, I mean, we don't just give out information. I mean, we go through all sorts of hoops. But because of the uniqueness of the services that we provide for the HIV patients, including the other ancillary or allied outpatient services, like renal dialysis and therapies and everything else, we get hit with a tremendous number of additional information requests.
DR. STARFIELD: Jerry, do you have anything to way in on that?
MR. HENRY: I think that our percentage is much greater when you consider the EOBs. Now, interestingly enough, when Medicare and HCFA decided that for the Medicare and secondary payer we needed to submit just value codes indicating what had been paid primarily, those dropped off and they have proved it can work. They will find during the audit process, if we didn't report the prior payments correctly. I think that is the way you are doing it right now.
But it seems to be a matter of trust between the payers and the providers that we report the prior payments correctly. As I said before, the information is on the claim form and it is not really needed. Now, we file Medicare secondary payer claims electronically today.
MR. MOORE: There is a separate requirement in the electronic or coordination benefits. I think that I have heard that some didn't want it included in the transaction set and they were mainly payers. It is going to be included in the transaction set and it is at the option of the provider not the payer.
MR. CHENG: It is that trust. We know where you are coming from. That is essentially it. Medicare is so big. We have dealt with them for so long. We know where they are coming from. We trust them. We really do. And they trust us, too, to an extent.
MR. MOORE: It looks like it when our attachments come out to you, doesn't it?
MR. CHENG: Except when you ask for an attachment, yes.
MR. SMITH: But generally we know what Medicare requires ahead of time. So, we have dealt with them enough to know that they do require for this procedure -- for instance, a real simple procedure, if a patient has a surgery procedure and has to have a chest x-ray, we know that Medicare is going to require a report on that. Why did they have a chest x-ray, before they will pay it. So, those types of things we know. But it is the other areas -- managed care particularly, who are developing new requirements almost on a daily basis.
MR. MOORE: One of the questions that has been asked -- I think Bill answered it -- a number of providers have asked the question of will all the managed care plans, including ERISA plans be required to develop the capability under these regulations to take electronic transactions. And the answer to that question is "yes," that they will not be allowed to make a provider submit it in paper.
If the provider chooses to submit it in paper, that would be allowed, but it is the provider's option.
MR. CHENG: No exceptions to the rule?
MR. MOORE: On those transactions, there would be no exceptions to the rule.
MR. CHENG: Regardless of size of the payer? I mean, even the small mom and pop shops?
MR. MOORE: The small payers would be given an extra year.
MR. CHENG: But they will have to comply?
MR. MOORE: Right.
DR. STARFIELD: Jerry.
MR. HENRY: One thing I wish this committee had asked this morning when the payers were here, you heard them talking about conditions under which they required attachments, but nobody thought to ask them had they ever told the providers what these rules were. Now, Medicare tells us -- we know that if we give them a 99 HICPIC code, we are going to have to explain what it was.
And the example, the 22 modifier they were talking about, if payers knew that, they would not send as many attachments as they do today. I can tell you our hospitals are the same as Paul's hospitals and what you are doing is that if the perception is they are going to ask for the attachment, we put it with it up front.
The ER report goes regardless because you are always going to get a request for it. Because we don't know what their rules are. We don't know what they are looking for.
DR. STARFIELD: Well, thanks very much. This has been an enormously helpful panel. I think you have given us directions that we might not have had before.
We are going to take a break for half an hour and come back at, let's say, 3:25.
[Brief recess.]
DR. STARFIELD: I would like to call on Saul Gelipko(?) from the National Capital Blue Cross/Blue Shield, who has a comment he would like to make.
MR. GELIPKO: Thank you. My name is Saul Gelipko. I am the manager of the Systems Analysis and Planning Department of Blue Cross/Blue Shield of the National Capital Area here in Washington. I have been here since 8:45 and I have listened to the fascinating discussion that has taken place this morning. But I have some concerns.
I am a payer and I have a concern about the rather conservative stance that the payers took this morning in basically saying, well, we need attachments and we need them and that is it and if we automate that is good, but we need them.
I am also concerned about what I find to be the rather -- well, let's say, somewhat hysterical reaction of the provider community in tying attachments as a code word for the payers are gumming up the works or they are finding a way not to pay claims.
Why am I so concerned? I am a payer who believes that there is no need for attachments. If there is a conspiracy, I am also a conspirator and what I conspire everyday is to find a way to pay claims with zero time spent on them. And as part of that, when it comes to attachments, I challenge our medical affairs staff. I challenge our claims processing staff everyday to tell me why do we need attachments because attachments do gum up the works. They gum it up for us. It costs more for us. It costs more for the provider community and it is not the best thing for our subscribers.
So, having gotten that out of my system, I would simply like to ask the committee to take the same approach that I feel that we do at Blue Cross/Blue Shield of the National Capital Area when it comes to attachments. Why do we need them? We start with the question, let's have zero attachments and make the case and make a strong case that there is a need for attachments. If you do that, standardize it, but let's not take for granted that we must have attachments, that there must be a thousand of them. Let's just get them into numbers, a 225, a 322X, that sort of thing. Make the case, please.
Thank you very much. I appreciate your time.
DR. STARFIELD: Thank you.
Are there any questions about that?
[There was no response.]
Thanks very much.
Okay. Steve Barr from HCFA is on now to talk about the Claims Attachment Project.
MR. BARR: Hi. Thank you very much for allowing me to speak to you --
DR. STARFIELD: Oh, there is one small change. We will adjourn at 4:45 today, since we are exactly on time starting, Steve.
MR. BARR: I am ready to go. That was enough.
I want to, again, thank you very much for allowing me to participate and speak to you today. I am Steven Barr. I work for the Health Care Finance Administration. I am the leader of the Proof of Concept Team working on attachments. I am the co-chairperson for the NUCC on attachments. I am a co-chair of the special interest group of HL7 working on attachments and I am a member of the 275 work group that is working on attachments.
So, I have no life. It is only attachments.
The other question is why did I do this? I guess the biggest question I would like to address before I go into my presentation is some of the things that you heard earlier in today's session from both payers and providers tend to make us very confused because, one, as a group of people from the -- working on the Proof of Concept Project to automate attachments, the HIPAA regulation just said automate attachments. It didn't say anything like what an attachment was, what automation was, what the whole concept was.
If you have heard the people speak earlier today, the attachment arena is probably the most confusing area of any area you could imagine. And we just took the charter and the words that were listed in the regulation to mean automate attachments. People have come and said eliminate attachments, set medical policy for attachments, direct payers to universally set medical policy.
That was not the charter of this group. We took it upon ourselves to take literally what was in the law regarding automation of attachments and to automate existing attachments. The issue regarding policy and elimination is a very, very interesting aspect. It is a project that should be done at some point of time, but I can tell you the Proof of Concept Team, and me as an individual, don't want to go down that road because I had enough arguments with this project. But it is very, very worthwhile and I think maybe at some point of time this panel should look into the elimination of all attachments, but we are going to go into the automation concept now that we have started to undertake.
The Proof of Concept Project started literally about two years ago when HCFA got together and started to ask payers and providers to tell us what they really wanted to do regarding automation of attachments and that was regarding the question and the response.
The first thing that they said was we would like you to look at the automation of the question that is asked from the payer to the provider. And HCFA, along with five contractors who were working on the Proof of Concept Project at that time took an undertaking to automate the question that was asked from the payer to the provider and developed a transaction called the 277, which allowed for the automation of all questions to be done from the payer to the provider.
That eliminated the need for a paper transaction to go out from the payer to the provider. So, what we did was did a cost savings from the aspect of the payer to the provider. We proved the concept worked. We developed the transaction. We developed the implementation guide and we thought we were done.
Then we thought we would go on it just a little bit more and start working on the response back electronically from the provider to the payer. In the middle of our entire project, we got hit by the HIPAA legislation, which not only mandated the attachment project, but gave us a time frame where we had to meet the deadline.
The same ten individuals who worked on the Proof of Concept Project for the 277 were the ten individuals who volunteered by twisting some arms to work on this project. The ten individuals -- and I would like to mention them because without them, we wouldn't be in the shape we are in now, whether it is good or bad.
It is Gale Carter and Conny Nichols from Anthem; Christine Stahlecker and Joanne Weingarth from Empire; Durwin Day and Maria Ward from Health Care Services Corporation; Julie Meehan and Mary Lynn Bushman from Highmark and Mike Cabral and Gerry Czupryna from United Health Care Corporation.
We started then to work on the automation of attachments and when we began, one of the first things we had to do was we had to make a decision amongst ourselves what an attachment was and how we were going to automate that. We took existing surveys that occurred from -- WEDI a survey. The NUCC did a survey. HCFA did two surveys to try to determine what are the most prevalent attachments that we should tackle first.
Of that list, we came up with -- from all the surveys and in addition to the surveys, we hand reviewed over 500 individual surveys. Some individuals who worked on Proof of Concept gathered all the attachments from their companies and literally hand-reviewed them and logged what kind of attachments there were.
So, we had data and we had some kind of idea of what it was. The only problem we found with all the surveys we did, the responses were very limited. A lot of people did not respond and a lot of payers did not respond, nor did a lot of providers. Either there was a lack of interest or they thought this was going nowhere fast.
But even without that input, we decided to go ahead and do this. We came up with 19 attachments and paired that 19 attachments down to the following 16 attachments. These are the 16 attachments that the Proof of Concept Team will tackle first when it comes to automation. I will read them to you.
It is ambulance, Home Health Cert. Form 485, Home Health Cert. Form 486, Home Health Cert. Late Lyme Disease, Home Health Cert. Blue Cross/Blue Shield of Illinois, outpatient rehabilitation, alcohol substance abuse, emergency room reports and records, notes/reports, referrals, explanation of benefits, workers comp, skilled nursing facility, auto liability, end stage renal disease, plan of treatment, invoice, letters, lab and pharmacy.
There are the 16 we have pared down and the 16 attachments we are planning to automate on the first pass. That is kind of an ambitious task when you consider that for each one of these kinds of attachments, there are a minimum of 50 versions and for each attachment, there are a minimum of about a hundred questions.
To pare those down to a workable project is kind of hard. What we did is we took all the data for a particular type of attachment. For example, ambulance, we took all the data elements from all the ambulance attachments we had gathered. So, if we had 20 different versions of ambulance, we took every question and if they were similar, we made it one question, but if it was different, we listed it down.
If you turn to the fourth attachment down, it is marked "data element," you will see 48 different data elements for ambulances. They were the current data elements that we found and only the current data elements that we found from the information that we had available. That does not mean there is not existing data that is out there, but this is all the data we can find from all the surveys that we got from all the attachments that were sent to us.
MS. FYFFE: This is a composite?
MR. BARR: Yes. There may be more data elements out there and there may be more attachments out there, but they were never sent to us and were never added to our input.
As you can see, some of the questions appear to be very, very well-written until you start trying to analyze them. As Clem will attest to -- and Clem was the brunt of all this because he had to work with us and he was another one who thought that attachments were just a simple thing -- when you start asking questions like patient's weight, for example, the question is is what kind of weight are you talking about. I mean, depending on what kind of weight you are asking depends upon how you want to pay the claim.
Is it stated? Is it estimated? Is it measured? Is it the actual weight? They are all various questions that relate to a patient's weight.
When it comes to distance, there is a question of actual distance; you know, what distance are you talking about? Just the question, distance, is very vague.
There is a lot of questions, like air ambulance justification, what does that really mean? I mean, there is a whole realm of these questions that although they are there are not very clear in what the definition of the questions are.
DR. LUMPKIN: It is up in the air.
MR. BARR: Yes. And it is up in the air to the point where the poor payer -- well, I say the "poor payer," the payer asks these questions and the provider has to guess what it is, you know, and if you are a provider and you get one of these attachment questions for an ambulance, you get nine different payers sending you questions and nine different ambulance attachments. You don't have any idea what they are asking for.
This variance of questions and data is the hardest thing to go through. In our first task when we came to these spreadsheets was to try to decipher what the question really meant. Well, that became problem No. 2 because these attachments were developed years ago. They were developed by payers for whatever reason no one knows.
The people who developed the questions have long gone. There is no documentation anywhere around. We have been told how you fill out this questionnaire. Well, we know what the answer is. We send it back.
MR. MOORE: HCFA didn't do these.
MR. BARR: Yes, HCFA did do this. HCFA is not escaping this part either.
But that is a major problem. I mean, ambulance may be clearer than some of the other questions, but this exists for every one of those attachments. There is no one there and there is very little procedure documented to ask where these questions are or what they relate to.
So, what we then undertook --
MR. MOORE: I asked the question of the payers, did they really review this information that came in on these attachments. And the answer, I think, was "yes." How could we review that if we can't even find the people who asked the questions?
MR. BARR: The people who asked the questions and wrote the questions down have long gone in most situations. It has just gone from word to mouth, word to mouth. The provider knows or makes an assumption that they know what the question really is.
If it is not right, you know, they know the next time to respond to something different. But one of the problems is is you have a group of people who are trying to streamline this process and there is no one to ask for documentation purposes.
So, what we did, we carried on in more ways than one, and what we did is we tried to streamline the process. The next set of questions are a spreadsheet that we developed, which we took the questions from the ambulance and tried to find a more definitive reason or what the reason really was or what the response code was or where the reference came from, whether it was an NSF, whether it was a UB-92, whether it was an 837, tried to find some point of context where it should be.
We did this for two reasons: One, to clarify what the question meant and, two, to have someplace of reference later on to decide whether it stays in the claim or it goes into the 275. This is a working tool that we looked at we could do later on. When it comes time to decipher where the data should be, we have a working tool where it is today.
This is reiteration No. 4 and as I can tell you, the people who worked on my project, if this doesn't work, put a fork in it. They are not doing it anymore.
This spreadsheet alone took staff time of one person three days to get medical review people to try to ask questions on this attachment. They had to sit with a medical review person for three days just to ask one attachment from one payer what the question meant.
Now, if you realize how many payers there are in the country and how many attachments there are, the time consumption to do this kind of project is over-laborious. And we have had to do it over and over again, once by Clem saying it was not right and then we decided it was not right and then Clem said it wasn't right. Then we did it over again. I have to put of the blame to Clem.
But we have gone this far and what is identified is is the question the way it exists today and you are saying, well, the question wasn't clear then. What makes it any clearer today? It doesn't, but at least we have a definition of what the question meant. And we do have documentation for every one of these data elements of what they really mean from specific payers.
Now, that is, you have to visualize that this is going to be done for all and has been done for all 16 attachments that we listed. There are two of them here that are No. 17 and 18. They were given to us by Bob Davis to work on at some point of time and we told Bob Davis that if we can prove the concept of automating attachments works, then this just fits in automatically to the process. But that is the reason why there is 17 and 18. He has already done some homework in advance of where we are going.
In addition to all this, which is no menial task from one year, we were chartered to begin doing and implementing and developing an implementation guide for the transaction that was chosen, which in itself was kind of unique because this is the first transaction that allowed the interaction between two standards bodies.
We are using the X12 275 patient record, which had to be completely changed, and HL7 data for clinical information encapsulated in the 275 for this transaction. Now, this is the first time that two standard bodies have worked jointly together in development of one standard. And they are both ANSI accredited and they have both worked very diligently with us in this whole entire process.
The implementation guide has been done, except for what they call the front matter. It is ready to go, thanks to some hard work by the group. We finally got a name. It took us two days to go at the ANSI meeting to get a name because no one would agree on our name.
And as somebody said, it must be idiots because you can't agree on one name for a transaction but we finally got a name done. After they changed it 13 times, we finally got one approved. Don't ask me what it is. I gave up after the 13th version. It is something to do with patient information --
PARTICIPANT: No, it does not have patient information.
MR. BARR: What does it have in it?
PARTICIPANT: Additional information to support a health care claim or encounter.
MR. BARR: Additional information to support a health care claim or encounter.
The 275 record as it originally was developed was a transaction that was done from patient to patient, provider to provider, data exchange. We had to change the transaction to meet our needs to make it from provider to payer. So, the transaction had to be changed and the entire implementation guide had to be written. That has been done and we are ready to begin the next phase.
The next phase is trying to get the codified data that is needed to ask the questions. If you look at all the questions that are there, and a lot of them are in free format text, a lot of them are nebulous, a lot of them are very vague, what we had to do is working jointly with HL7, develop a method to do the codified information and we developed was a current messaging standard that was given to us and suggested to us by both X12 and HL7, which was a message called LOINC, which is logical observation identifier names and codes.
This LOINC message or HL7 messaging standard is very prevalent within the hospital-based community. People are correct. It is not presently prevalent in the physician-based community, but that is today. If you look at it today and you implemented this transaction today, you would have very little participation from the community that is out there today.
But this transaction is not geared for what exists today. This transaction is geared for the year 2001 and in 2001, we hope that HIPAA will already have an impact, that all the providers in this country will be automated, using an 837 for the claim and we hope there will be a very big trend within the entire provider community to automate their systems. The trend in this country is heading towards automation in provider systems. It is not very, very fast, but it is steadily growing as we proceed month after month after month.
LOINC messaging will allow, if done correctly, for the payer to ask a question by means of a LOINC code and allow for the provider, if he has an automated system, to have his extracted automatically from the management system. It never has to touch human hands and human intervention.
The question will be sent electronically from the payer to the provider via 277. The LOINC message will be encapsulated in the 277 asking the question. The response will come back via a 275 with the HL7 message repeated, so there is no question about what the message is and it will be sent back electronically to the payer.
It never has to touch human hands if it is done correctly. Now, we understand that this process may not be accepted readily or very quickly, but if you look at the number of claims that have attachments today and the number of claims that are accepted, if you take roughly that -- you have to give an assumption that there is more than 200 million claims and that is probably minimal because HCFA does -- Medicare does 80 million itself and I think you have heard today that some of the payers here do more than a hundred million, if you take 200 million as a base figure, which is probably on the lowest side you can get, figuring 20 percent exceptions, you know, that is 40 million claims that have attachments.
If you can automate 5 percent right off the git-go, you know, what are your cost savings. And that is only talking about 5 percent. We believe that the process will transcend itself as time goes on for this total automation. It is very -- you know, people are saying no one uses HL7 today in the physician-based facility. Well, that might exist, but we made a statement in the HIPAA regulation for the claim to tell people that if you are buying a translator to meet HIPAA mandates, you ought to find one that is HL7 compatible, so you only have to incur the cost once because HL7 data is coming out for attachments.
So, we gave everybody a pre-up, so if they go to have any costs, they can't say you never told us. Now, I will have to go and buy another translator. Yes, translators are expensive, but we gave everybody an option on the very front end that they can do this because we knew this is the way we were going to go.
What we tried to do when we developed this question is tried to make this process a little bit easier to work with. And we don't have a slide for this because this was just decided on Thursday in Los Angeles how we were going to do this.
Clem will probably be interested in this unless you heard it from somebody else already. What we decided to do was take everyone of the 16 attachments that have been decided an individual LOINC code for the attachments. So, ambulance would have its own LOINC code. Within that LOINC code, all the data elements for all the attachments would be another list. So, if you had 400 data elements, they would be listed in a separate list, which would be List No. 2, which would list all the data elements with a LOINC code and the message the LOINC code represents.
So, you have a list of all the attachments with the LOINC code, a list of all the data elements with the LOINC code and the third table would have the attachment with all the data elements contained in that attachment and the applicable LOINC codes that pertain to it.
So, there would never be a question and it would be cross referenced all the way around. We thought that was the best way to work this out. It would allow for a lot easier transition if we had to make changes because they wouldn't be imbedded in a transaction. They would be outside of the transaction base. So, changes could be done outside of the transaction.
They would be maintained by the X12 Reason Code Committee. That is where the references point is. The data maintenance to that is still an issue with our group. We don't know how to do the data maintenance. We haven't addressed that data maintenance issue that was brought up earlier by members of this committee of how we perceived doing the data maintenance, but that is an issue we have to tackle at some point of time.
We are looking for data maintenance to be done in hopes of having the NUCC, NUBC or WEDI being the governing body for this data and having the expertise to do it as the data maintenance would come along, just like they would do the data maintenance for the claim. So, it would be one governing body doing the same data maintenance and you wouldn't be going from place to place if you had a request.
Where we are now? We are this close or this far to succeeding. We are beginning -- we are scheduled to begin testing of our standard that we are doing with the 275 and the HL7, to begin testing at five payer locations in this country, which would be the participants that are working on this project and commercial side for each one of those in April -- I believe it is April or March of 1998.
Once we begin the testing, we will have data analysis done. We are looking for a federal agency to help us with the data analysis, so it cannot be said that we skewed the data. It would be done outside of our purview. Someone else would maintain that data. We have already asked, I think, CDC or NCVHS, NCHS, to provide us statisticians to work on this project. Once we give them the data, they will assimilate the data, so it won't be considered that the Proof of Concept Team skewed the data in any means or fashion.
We will prepare an executive summary, which will be done by HCFA, which would be available to anybody to show what the Proof of Concept did. And that is not all. We are going to take another task on -- that was mentioned by Dr. Lumpkin in the last NCVHS hearings. Dr. Lumpkin mentioned it, but we had already looked at it in some other aspect and had a special meeting in Chicago the week after the last NCVHS hearings.
This is probably the biggest bang that we are going to come across and the biggest task that we are going to undertake. What we have tried to do is tried to make an awareness of payer and provider in this country to begin standardizing the questions that are asked from the payer to the provider. What we want this group to do is we will take and give them the attachment questionnaire and tell them to go back and look at the questions and start standardizing how the questions are asked and whether the question is correct or what should be the true meaning of the question.
Then we will do that for all 16 attachments. The people who are working on that, who are volunteering for that, are -- we will show you the list in a second. There are two sheets. But these are some of the people -- this is one sheet of the people who have volunteered to do this and there is another sheet of people who have volunteered to do this.
There are payers and providers doing this and what we hope to do is have them work independently of our project to do this. We don't want to be part of this task. We are on the work group, but we don't want to be the leader of this project. We have asked the NUCC to consider the factors, would they consider taking on the leadership role of this responsibility for this project. We would like the NCVHS to consider another -- any alternative group that it would be the responsibility.
People have said that if HCFA takes the lead on this project, becomes a leading focus on this, then it has a HCFA or Medicare mandate versus a universal mandate. The standardization of the questions and the standardization of attachments is not a HCFA mandate. It is a universal mandate. And we have gone beyond a narrow focus. We have gone to a wider focus.
These questions that will be done will be done outside of the purview of the attachment process. We know that these people cannot get the questions resolved and defined before we have to write the regulation in 2-98, but they can begin working and as they begin working, we can start incorporating it, encapsulating it into the transaction.
So, there will be an ongoing process. This is strictly volunteers. Everyone here has volunteered to do this. We look at it as probably one of the -- as mentioned by Dr. Lumpkin before, the biggest task ahead of anybody is to start standardizing the questions. Again, no one wanted to tackle this project. We have got the ball rolling. We have got an organization meeting. There is a meeting scheduled for February 17th, which will be the second conference call we will have. You don't know it yet, Clem. I don't think you have checked your schedule -- February 17th from 12:00 -- I think it is 12:00 to 1:30. You had a meeting after that -- 12:00 to 1:30.
Clem will run us through the process and then, hopefully, hand it on to the work group.
DR. MC DONALD: I missed one meeting.
MR. BARR: Yes, you missed one meeting. You get tasks, you know. Clem will try to explain it to the people. We had a cursory phone call two weeks ago, but it was just very, very haphazard. We didn't have enough lines. People were notified at the last minute.
This we have done in advance. People will be notified. We have given everybody on here everybody else's e-mail address, everybody else's -- who they are. They have got this exact piece of information. We are going to provide them with all these spreadsheets prior to the telephone conference.
So, we have a working paper to work with. In addition to all this, there is another group that has formed that has the horrible task of writing the regulation for this project. Now, one of the problems is it is not like the claim, where there were hundreds of people anxiously biting at the bit to find out about the claim and everywhere you went, all people wanted to know was about the claim, the claim, the claim.
In this process, there is only a small handful of people who know this entire process and it is very difficult to try to transcend the complexity of this project from one to the other, but there is a HIPAA implementation team. I am the co-chair along Phyllis Delavaris(?) from NCVHS --
MS. GREENBERG: NCHS.
MR. BARR: NCHS. Excuse me. NCHS, who is working with me. There are other members from other federal agencies who are also on the HIPAA implementation team. And we are chartered with having an NPRN out by 8-98. Rather ambitious, but -- now, that is not all.
You think that is enough, but it is not. There is also -- it was mentioned earlier by Bob Davis that did another group that is meeting to start standardizing questions from components that aren't in the mix. And that was done by NCHS, which Marjorie has, I think, organized. I will be a part of that because we want to make sure that the questions that come to us are not vague, are clear. There is a definition and it only has to be sent to us once and one time only.
Their undertaking is a very, very tremendous help to us because they can give us one set of data and one set of data only for whatever they need. Then it is very simple to do what we have to do to get it into place. We would rather have one set of data and a hundred questions, rather than 50 different groups pointing at us and saying here are two data elements. Figure out are they repetitive or not.
What we did also and what we are working on at the last meeting, there has been a joint work group formed at the X12 meeting, which will allow the interaction between the 837 claim and the 275 work group for attachments, which will try to begin this migration of data from the 837 to the 275 and from data that was considered as part of the 275 to the 837.
This group will meet every X12 meeting on the Monday of the meeting for two hours to begin some migration plan. There is also monthly teleconferences scheduled for this group for two hours every month to begin the process going. So, we can start looking at the transaction of the 837 and to see what the data element is and see if it belongs in the claim or belongs in the 275. And we can have some discussion and begin the discussion process and develop some kind of migration plan.
In addition we are having open forums at every X12 meeting. We had one at this time just to discuss the data correction activities of the 275 and whether the 275 should be used for data correction of a claim. We are going to go back to revisit that issue again at the next open forum for two hours to discuss it.
When you discuss a claim and you discuss the 275, people come out of the woodwork to hear what is going on. When you discuss the attachment, it is very, very low key and people really don't understand the impact. It is my opinion and probably the Proof of Concept Team's opinion that if the claim was seen to be an insurmountable task and was done, then attachments are the Mt. Everest of insurmountable tasks.
There are only ten of us doing it and they work on other projects. They don't work full time on this project. Their time and resources are very limited and I am very pleased at the fact that they have given me so much time. But, again, we are being stretched to the very limit of our resources and we are going to go on from there.
That is all we have done.
DR. LUMPKIN: I was looking for surgery on your list of attachment types and I assume that is under 5, "notes and reports."
MR. BARR: Yes.
DR. LUMPKIN: What kind of questions are in surgery? Because it seems that the kind of information there -- I mean, that seems to be the most -- potentially the most amorphous area when people just want to know what you did and why you did it.
MR. BARR: What we were going to do on the surgery notes is that if you need the entire surgery note, you can ask for one attachment, which would be the surgery note, whichever the LOINC code is --
DR. LUMPKIN: So, it is an attachment to the attachment.
MR. BARR: No, it would be one attachment, which will be surgery and you can ask for the entire set of surgery notes. If the surgery notes are already in an automated version or an automated style by utilization of a LOINC code, you can go into the system and pull out the entire surgery notes that are already transcribed, download them into the 275, into the bin segment of the 275, which will identify that as a note section and send it.
If you don't have transcription and you have the capability of data entering through ASCII text, you can resubmit it again through ASCII text. The question becomes is what kind of -- surgery notes are usually very -- they usually want everything. So, that we would give them as everything. If there is certain specific data associated with surgery notes, they would have special data elements contained with it.
DR. MC DONALD: There actually is a format from the Joint Commission for a number of headers in the surgery notes, blood loss, NSCVUs(?) and a pre-op diagnosis, post-op diagnosis, surgical systems, that kind of thing. It is imaginable that someone would want to sub-select on those things.
DR. LUMPKIN: I kind of have two additional questions.
DR. COHN: Well, go ahead. There is so much to ask here.
DR. LUMPKIN: There is. I am fascinated by these LOINC codes. Of course, my first thought that is if you could just write them all down on a silk scroll, you would have LOINC cloth.
Are you envisioning these as being a -- if you are going to ask information about an ambulance that you have to ask, you send all of the codes, the 40 some odd --
MR. BARR: No.
DR. LUMPKIN: -- or they could pick and choose. Could they pick and choose one -- a couple of codes out of one and a couple of codes out of five and six and essentially ala carte make up their own transaction?
MR. BARR: You could. I mean, how we visualize this transaction, if you need the entire attachment -- if you need the entire ambulance record, there would be one LOINC code to identify the entire ambulance record. That would be LOINC code for simplicity purposes. It would 00001. If you would request on a 277 saying I need the ambulance record, LOINC Code 00001, it would go to the payer and they would know 00001 means the entire record.
If you need certain data sets or data elements contained in that, what you would say is -- and the transaction 275 allows this -- you would say I need ambulance attachment LOINC code 00001, but I also need Question 1 or LOINC 0002, 0006, 0008, 0009, 010. So, in that situation, you would need only five questions rather than the entire attachment.
So, if you need specific data related to that, you can then go and ask that specific questions.
DR. LUMPKIN: Then, finally, did you envision this list of 16 as being that is it?
MR. BARR: No. We didn't get into that. We visualize -- as I said before, the people who work on this don't have a life and in our afterlives, after this part goes, whether they like it or not and, hopefully, they will like it, they will continue on with the next phase. Our next phase will involve -- we have defined the 16. We know the concept works. We have gone through and made any necessary changes to the implementation guide. What else has to be done?
We know, for example, Bob Davis gave us two from New York State that have already been defined. We have Medicaid that hasn't been addressed yet. We have got the public area that hasn't been addressed that. Whatever other group, whatever other attachment, these are the 16 most commonly used attachments that we determined. There may be 400 more out there and that is just shaking our head.
Like Clem said, the questions are infinitesimal of what is out there today. We don't know everything, but these are the 16 that are most commonly used today.
After this, they would say to us I think you should address the next issue, which is some other attachment or you need these data elements. We would get guidance from one of the other groups.
DR. MC DONALD: Just a little bit of clarification.
Some of these attachments are bigger than they look. Like "lab" includes any kind of lab tests in the universe. So, thinking about that as one attachment is not quite the same as saying the ambulance one, which is sort of a predefined form.
And some of these other ones, "notes/reports," that is a very large space.
DR. COHN: First of all, Clem, I am very aware of how big all of this is. I mean, I have no question about it. I am not going to express an opinion. I just want to start with a couple questions. And I think the question as I look at your 16 types is I was actually just -- just to get very specific on a couple of things, can you explain to me the difference between the referrals attachment versus the X12 referral standard and how they differ in terms of what they do? And similarly, under "workers comp" on 8, workers comp attachment versus the first report of injury standard that is in X12, I mean, are they truly different items?
I mean, are we having -- I am somewhat reminded somewhat jokingly that the reason I love standards so much is because there are so many of them. Are they truly different beasts or are these just different ways of doing the same thing?
MS. STAHLECKER: I am Chris Stahlecker with Empire Blue Cross/Blue Shield.
Your first question on referrals, there is an existing standard transaction set to deal with the referrals. That is the 278. But it did come up for analysis within our work group. So, we did want to have a look at it and fully understand the work flow involved around referrals. So, at this point, it is on our list. We are not assured yet that we do need to include it in a test of a Proof of Concept. We are not sure if it is a supplementary piece of paper that is the copy of the referral back to the primary care physician from the authorizing payer organization or if it is something as simple as the referral number that was released by the payer organization.
So, we are not quite sure yet the scope that is involved with referrals. And then workers compensation, in New York State there is a form. It is called the C4 form and it was submitted to us. It may fall down -- Item No. 8 may fall down with 17 and 18, but it was referred to Empire for analysis as we did our data collection and our outreach to some of the organizations that had attachment types.
DR. COHN: Now, similarly, that is also -- let me ask about No. 16, which is pharmacy. Is this the NCPDP standard pharmacy transaction that is encapsulated as an attachment or what pharmacy is this that we are talking about.
MS. STAHLECKER: Again, at Empire, we had a working model actually of collecting lab and pharmacy information. The working model happens to be with a firm that is called Quest Diagnostics. And they are actually submitting lab test results to Empire for input into an early disease detection program, where we are actually analyzing the results for a couple of purposes. One is to make sure that what is billed on the claim was actually performed, but another one is to make sure that the actual results of the lab test is going to result in improved care for our contract holders.
Since we have a lot of information already on file about history of other types of lab tests that were performed, our specialists, clinicians and physicians can do some analysis on that. So, the lab and pharmacy kind of go hand in glove. With our test provider, New York Hospital, we did want to receive in a test mode both lab results and pharmacy data.
The pharmacy, we are still analyzing whether or not that is going to be pharmacy orders only or if it can be filled pharmacy statements because some of the attachments that we do request include methods of administration on the actual drugs that were dispensed.
So, this is part of our data analysis at this point.
DR. MC DONALD: I think it is HL7.
MS. STAHLECKER: Did I answer your question? Perhaps you are asking more what standard are we going to be using?
DR. COHN: Well, certainly, as well as how this is differentiated from -- I mean, there are certain things that the committee has already recommended, which I think are going to be coming out in those likely national standards. Assuming we see notices of proposed rulemaking and all this, I was just trying to figure out where if we were duplicating some of those standards. So, that was really the question in trying to understand how these were different.
MS. STAHLECKER: So, for referrals and workers comp, we may drop them to the bottom of the list, but the lab and pharmacy, we did expect to receive an HL7 standard formats embodied in the envelope of the X12 275.
DR. LUMPKIN: As I understand, our pharmacy transaction is really between the pharmacy and the payer. What I understand this is --
DR. COHN: It is a claim.
DR. LUMPKIN: It is a claim. And what I understand this is is that, for instance, let's take a hypothetical instance of an obstetrician, who gets a call from a managed care firm and says you wanted the patient to stay in the hospital a day later. Why? Were there antibiotics given? Were there IV? This would then be "yes," I gave this antibiotic. I gave it this way for that patient. So, it would be more the clinical side of the kinds of questions that are currently being asked to which there is clinical information requested, which is different than a claim.
DR. COHN: You might be right, but it is clear looking at this list. I mean, that could be also a report.
DR. LUMPKIN: Well, that is because there is only one word there.
DR. COHN: Well, I mean, that is -- I guess I am left with what is -- okay. That was my question.
DR. MC DONALD: I think that one of the common claims and it is almost always directed at hospitals is give me all your drug orders as an attachment sort of a thing and then they look at it to see whether they got -- why were they getting IV antibiotics, why they are not getting any IV antibiotics and they are still in the -- I don't know what they look at it for. We are going to learn all that later.
That is one of the kind of things you see asked for.
MS. FYFFE: In looking at the list of attachments and also the types of questions that go along with ambulance, when I have my fraud and abuse hat on, I can completely understand why these attachments were chosen and why these questions are being asked.
Have you talked to the Medicare integrity people or the anti-fraud people in HHS because if, in fact, over time you can become more efficient in collecting this information, I think that you will see the problems with fraud and abuse, hopefully, become reduced over time. I mean, whenever you go up to Capitol Hill and there is testimony about fraud and abuse, they talk about home health. They talk about ambulance. They talk about substance abuse in psychiatric hospitals.
I just want to stir things up here a little bit.
DR. MC DONALD: Well, these questions, I think, came off -- again, I didn't do the original development work -- off of real attachments that people are asking for.
MS. FYFFE: Yes, because there is a problem.
DR. MC DONALD: There are questions about many of them. I mean, the thing about asking about the patient's weight, I have never -- we got 110,000 admissions this year and I don't think I have ever seen the ambulance drivers run and pick a patient up and weigh them, ever. So, I don't know where they are getting the weight from and I don't know what they are using it for. Maybe they charge by the pound.
[Laughter.]
MS. FYFFE: The question that comes to my mind is was there a high tech neonatology equipment in the ambulance that you are charging for. And if you have a patient that is 200 pounds, why are you charging that? Okay? If you have a patient that is 400 pounds, you have to have a special gurney or something on the ambulance. I mean, you get into all of these questions about why are they charging this exotic stuff, well, for a patient that weighs, what, 32 grams or -- I mean, it all makes sense when you are a payer.
MS. WARD: My name is Maria Ward. I am with Blue Cross and Blue Shield of Illinois and I am a member of the project team, the Attachment Project Team.
In response to your question, Kathleen, last year we met at HCFA Central Office with -- his name escapes me -- Kevin something, medical review. Our understanding was that he was over at least part of the Medical Review Policy Division within HCFA, to bring these issues up with him in terms of the HCFA perspective on things like fraud and abuse and program integrity and things like that.
So, yes, we did do a little bit of a outreach there to include them.
DR. MC DONALD: Could I give another cut at -- because I have been -- I may be the guilty party in some of this evolution. Firstly, the LOINC codes are just a convenience. We just need some codes and what those evolve into or would be, that is a totally separate thing. And you can think of what the data being sent is mostly being sent
-- you are sending a spread sheet and you can think of it as an estimated 12 or a CBC and you are going to either ask for the whole thing or you are going to ask for the parts. There are names for each of them.
Now, there is some slight inaccuracies in how this is drawn is that you wouldn't have a separate field for units when you -- it would be part of the row in the spreadsheet. So, in that spreadsheet, each row represents a single observation, which would include information as pulled out of the local system, which would usually be things like units and normal ranges as well that come along with it.
This is a fairly flexible structure that they can embellish and expand. The challenge is in naming these damn terms, naming them first, because adding the codes to them is just mechanical. If you look at things like distance, that is the name, well, that is not going to stand alone very well. Even in an ambulance report, is that the distance from the door -- the gurney was rolled? Is it the distance they drove, you know, at a hundred miles an hour? I mean, there is about a hundred ways -- so, there needs to be much -- when we finish this, we need to make those names much more thorough, so they stand alone.
Then the other thing is when you look across different kinds of attachments, you are going to see similar looking things, which might really be the same thing and we ought to get them, you know, lined up, so they are. That is where the tremendous work in this activity is going to be, getting those names finally made.
DR. COHN: Let me first of all make a comment. I feel, of course, privileged to know someone who has worked a little bit on LOINC. I look to my right here, so, I -- obviously, I am delighted to see that it is involved so closely in this project.
However, having said that, effectively the whole attachment standard, since it is so inextricably linked with this code set, I think that at some point we are going to have to sit down and look very specifically at that code set and the plans for maintenance, improvement, sustaining that code set, which is something I just don't know about.
DR. MC DONALD: Well, that is why I deliberately said that we use it as a convenience. We can deal with these things. We have tools. But it doesn't have to stay in that body or in those -- I mean, hopefully, the codes wouldn't change just for fun. But these are not -- we have all the lab codes in there. So, that is sort of started and a lot of -- some of the nursing variables that they are interested in for the nursing homes. That is not the key.
The key is that you are really talking about a table that you can -- you define as a live table, which someone would maintain. Hopefully, we would make that table as small as possible for any given purpose, which is another exercise of considerable effort.
Then, hopefully, we will get the names clear enough that they stand -- you look at them and you know what they are talking about. That is a huge challenge. I mean, if you look at some of these things, Transport 2-4(?) -- now, the ones with answer lists are helped by that.
DR. COHN: Yes, and don't in any way -- I mean, I am well aware of the gravity of all this. I am just sort of just identifying that we needed this to sort of check that as something we need to do to listen to --
MR. BARR: That is a question that we have because we really don't know who should do the data maintenance. We would suggest one of the standard bodies would do it, but we don't know who is going to do that. I mean, that is a task that has to be done, has to be assigned to some body and we are looking for some advice as to who should be the applicable party to do it.
DR. COHN: Right. And I think that is something you really do need to talk about as a separate topic.
Now, having said that, I am -- I haven't seen this presentation before. I think I would be a little less stunned if I had been going to the meetings.
MR. BARR: You are not the only one.
DR. COHN: And this standard is unlike most of the other standards in the sense that the other standards have been worked on for years and have gone through a number of iterations and, obviously, I am just a little concerned -- not terribly concerned, but a little concerned about both the scope, as well as the time frame you are describing. And I am trying to get a sense from this of what in your mind -- I mean, between now and June, which is four months, what reasonably -- I mean, what would be a reasonable value demonstration proof of concept model that at that point we could say "yes," that is something that really does work and that could be in an NPRN versus what could -- I am just sort of -- where will the phase 1, where the line is on this one. Do you expect to have demonstrated all of these things in large volume capabilities, in multiple different environments with a large hoard of people saying -- cadre of people saying "yes," this is great between now and the time to get the NPRN out in June?
MR. BARR: What we have done is gone out and each of the five payers that are involved, the Medicare and the commercial side, had the opportunity to select one of these attachments or multiple of these attachments or some form of these numbers that they thought they would like to work on considering the scope of their business.
So, if they had a payer that did home health, they would have the expertise to do home health. If they had an expertise to do workers comp or that was chosen, that payer would do workers comp. All 16 of these have been divvied up per se and have been divided amongst the five payers. Some have three transactions; some have four; some have five. There is an overlap that they all will do notes, things of that nature, but they all have been divvied up.
We hope that -- our biggest challenge was to get, one, the implementation guide written and have the applicable HL7 references that refer to where the location of this stuff, this information is, encapsulated in the implementation guide.
We have done the 275 X12 implementation guide. The only part that is vacant or has not been done is the encapsulation of the HL7 terminology and the HL7 references. We have turned that over to people with better knowledge of HL7 than the ten of us have and we have been told that they guarantee to us that we will have appropriate information available by March to encapsulate in the implementation guide. By March, if the creek don't rise and we don't get torn away from some other aspect of this, we will begin testing at the five contractor locations.
So, we visualized going through August originally. We hoped to test for six months. We kind of got short circuited by a mandate for us to write -- the HIPAA implementation team to write the reg by 8-98, which kind of short circuited our process by two months.
We figure if we start testing in March, we will have March, April, May and June, four months of data. Technically, we give this to a statistician. The statistician will tell us how many of each type of claim we need to satisfy the process to prove if it works or doesn't work.
They will provide us with that kind of information to say that for this kind of attachment you need 50 claims or you need a hundred claims because once you prove the concept works and you ask one question, conceivably the same question can be asked -- if you ask Question 1 and the process works to get the question, you can ask Question 2.
So, we will have that data done. That information will be given to us by a statistician, who will tell us how many claims each one of these need. We will give that to the work group. The work group will know for home health certification I need 75 claims to do this. They will enter 75 claims to meet the satisfactory limits.
All that data and the CONNA(?) claim and the CONNA results and all that will be sent, hopefully, Marjorie,
to --
DR. COHN: Can I just make a comment? I am very supportive of what you are doing. Do not get any comments wrong. I am just concerned that you are attempting to do a proof of concept and for the people on the Internet, I am raising my hands in a very large expanse; whereas, I would be as happy with something very well done that has a slightly narrower focus.
DR. LUMPKIN: And now his hands are small.
DR. COHN: Now my hands are small.
And that was really done very, very well. So, once again, you may be able to do everything with a very, very high level of quality. So, if that is the case, please proceed, but I think the quality is the key issue here, as opposed to the quantity. That would just be how I would judge a first version of this.
MR. BARR: It is our belief that if we prove the concept works with the question and answer and you can use the 275 transaction, which has not been tested, never been used in anybody's facility, encapsulated in HL7 together in one transaction, if we prove the concept works and the concept can be transmitted and if we can get this data flowing, in reality, all it is is putting other data in. The concept is very, very clear of how you do it.
The transaction itself is not as big as the 837. As a matter of fact, I have a copy. I will show you the size of the transaction. This is the size of the transaction. Now, this doesn't have HL7 data encapsulated into it, but if you look at this transaction, it is relatively very, very small, very concise.
DR. MC DONALD: I would just comment on its history. The 275 was actually invented to send encapsulated data, specifically HL7, and I was actually on that committee at that time, as were some other people from HL7. Then it just went into limbo, I guess, until this thing.
MR. BARR: I made a mistake. This is the entire extent of it. I only gave you one portion. This is the entire extent. It is a little bit thicker.
DR. MC DONALD: I think -- I mean, I have to share -- you know, having built systems, there is tremendous tough stuff that is going to come up everywhere. I don't want to -- I could give you a litany. So, this is not going to roll over and be a healthy, immediate, kicking and running child.
DR. STARFIELD: Bob Mayes.
MR. MAYES: Yes, this is Bob Mayes from Health Care Financing Administration.
Simon, let me see if I can propose a view of this and, Steve, if I am completely off base, or Clem, feel free to jump in, but I am seeing the world now ever and ever in terms of data registries and the registry approach. So, let me see if I can phrase how I could conceive what we are doing here.
A data registry basically just allows you to define medi-data about data in a standard way and then have a framework in which you can compare different sets of data that have been such defined. In a way, what we have here, if you look at each of the attachments as a data set, we are saying, okay, we want to define, first of all, how do you represent those questions? How do you represent the data elements?
That can be done by a wide variety of groups and, in fact, in reality, probably will. I doubt very much it is going to be NUCC or NUBC, who makes the data decisions on all of these different attachments, particularly those that were put forth by New York that are really more public health type attachments.
So, you might, in fact, have a number of what are called in the registry world, registration authorities, each group of whom has been given the authority to be the definitive group to define the data. They are going to represent their data needs in a standard fashion.
Then you have a mapping issue here and really what you are mapping to is two different things. You are mapping one to the 275, which since it really just is an envelope and has this big chunk is a fairly simple mapping, although there are some issues when you sit down and look at the different segments that you have to be concerned about.
The more detailed mapping has to be done to the HL7 constructs that you are actually going to use to represent that and HL7 itself is working on moving towards a model-based approach to how they define and how they build their standards.
So, in some respects, if we build -- and I think this is what Steve is talking about -- if we build that infrastructure, if you will, that registry infrastructure, to say, okay, we have got a group -- we have got a 275 here, if you will, that is a high level model. We have a more intermediate model, which is the HL7 constructs, and then we have these separate data sets that we are going to map to there. Each of those data sets is going to be defined at the implementation level on the ground by whatever the appropriate group would be, whether it is a billing group, whether it is a public health group or whatever.
Once you have that infrastructure in place and that process for doing that mapping, whether you do one, whether you do a thousand, it really doesn't matter. Now, one of the nice things about building that infrastructure in that way is it begins to then make it a little bit easier to see, well, gee, you know, the questions that this group asked are really fundamental. This gets to the point of, you know, weight over here really is the same as weight over here. So, let's see if we can't make sure that we have got some normalization across these attachments.
But they are always going to be very specific issues. There are issues about ambulances that don't apply to ESRD or to other things.
I don't know, Steve, if that complicates it, but in my mind that is how I am sort of conceiving it, as the regulation is really building this infrastructure, this framework, this process, whatever you want to call it, and it is going to be an ongoing effort in terms of actually quantifying specific attachment requests or maybe actually finding out that these have collapsed or completely gotten away from.
DR. COHN: Okay. Why don't you stay there for a second because maybe I am confused here. Also, let me just query the group about what the standard that we are dealing with is. Now, at one level we are dealing with the 275 envelope standard, but then we are also dealing with standards on how an ambulance attachment is and how a standard about Form 45 and on and on and on.
Now, I am presuming that our -- the NPRN that would be sent out would not only define what a 275 envelope standard is, but also go into somewhat exquisite detail on how all of these separate pieces are, along with the upkeep, improvement, maintenance, data model, clarifying, resolving all data elements as part of that, all of that. That is part and parcel of what that NPRN will be.
MR. MAYES: Well, Simon, though, but I would -- that is not quite -- if you looked at how we did the transaction standards, we just said, no, you are using 837 and then we defined what groups would be responsible for the ongoing maintenance, enhancements and changes to that standard. So, in fact, the existing approach with the other standards is not to put in regulation -- in fact, it was explicitly not approached that way -- was not to put in regulation every data element and say that you now have to change the regulation.
What we did is said "no," there is a process. There is an SDO that has this standard. There is a group that has been identified to have the content and no more often than once a year, they can change their mind and decide that they want to add, modify, delete.
It will run through the approval or review by the department, but we haven't taken that approach with any of the other standards and I am not sure why we would take that approach with this particular standard, particularly with this standard, since ultimately there could be a wide variety of specific contents of attachments.
DR. MC DONALD: I think it is simpler than what you described, at least how we conceptualized it, and it is not quite how it is here yet, but Wes Rochelle(?) is working on this slightly simpler version of it in HL7. All we really have is we have questions and we have sets of questions. And the sets of questions -- what we really start with are sets of questions because we do have these -- some formalized one page documents with specific questions on them.
There is a little bit of tweaking on those boxes on the sets of questions, which I don't want to get into, but like dates and values will be matched within a single question in this newer version. You wouldn't have to explicitly say also I want to all the data of the glucose. It just kind of comes with the glucose.
Then what we have the problem of is defining the questions, which is the fundamental problem anyway. I mean, that is the fundamental problem. And I mean, ideally, we would have -- and really if we look at the attachments, which are defined as form, there are probably only eight of them. There are these models to work off of and with a lot of just labor, we can make up -- we can get to a point where we have fairly rigorous -- well, the things that have been criticized we have to clear. Then you have the other blobby(?) ones, like "notes," and so someone has to decide how we categorize those notes and those become separate questions. If we want to distinguish by person and by, you know, whether it is the nurse's note or the doctor's note or the inpatient note or the outpatient note, that is what you see actually, the ER note, the admission note. Those are the kind of things you see.
Then we get into labs and there is other kind of diagnostic things that are sort of -- just people want to have, the glucose and the creatinine or they want to have the EPO dose, whatever those -- those are codes -- LOINC has codes for the labs and there are pharmacy codes in other places and so that we can get to there.
Now, pharmacy -- I mean, not tomorrow, but we could make a leap -- I don't know all 16 -- I think it is feasible but it is terrifying. And then the other -- there are problems connecting to the real clinical system to the horrible ones, you know. So, they got a word processing unit with just the patient's name in it maybe. I mean, you know, that is all they got. How are they going to get from this request to that in a useful way?
You know all those kind of problems, but I don't want to ponder on all the kind of problems, but I think the model is really just a list of questions. I mean, it is a master list of questions and then sets of questions that would correspond to specifically named attachments.
DR. STARFIELD: You started off by saying you started this process before HIPAA.
MR. BARR: Yes.
DR. STARFIELD: So, in fact, it is not -- what is not clear to me is how relevant this is to the specific HIPAA requirements. Now, it sounds to me that what you are doing is describing the way we would go about developing standards, not the standards themselves. Do I have that right?
MR. MOORE: One of the things that when we first started this, I think it was recognized in the legislation that the claims adjustment issue was not going to be as simple as the other transaction sets, first of all.
Two, we had a mandate to get this done. What we are trying to give is a status report on where this work group is. There will be a regulation that will be passed around the department through the data council to this committee, like all the others. But what we are trying to do is work on getting this documented as well as we can with some direction as to what is an attachment, what are the first pieces we are going to cut off and go after. And it was based upon the best information that we got back from the community that is providing that information and ask them for it.
So, you know, I think we are getting into such detail that I don't understand what Clem is saying over there when we get into the data sets and all that. You know, really, I think, we need to back off from the work group and let them do their work.
If we don't like what they are going to do, I think you will get a chance to say "no." But the department has got to put something on the table come this fall that says here is what we think attachments are. Here is our first iteration at -- whether we are going to take this much of it is -- does it represent 20 percent or 30 percent of what is out there? Maybe we can put an estimate on it. Then we are going to put out an NPRN that says this is what it looks like. If the community doesn't like it, it will never get adopted.
So, you know, your guidance that you can give us that says, you know, we need to be turned a little bit if we are going in the wrong direction, I think that is important. But I think we are getting into detail here on this one that is far more than some of us understand or --
DR. BRAITHWAITE: Let me remind everybody that standard setting is an evolutionary process that never ends. We are not here attempting to write a regulation next year that sets the standard for attachments that forever remains the same. We are trying to take at least a piece of this problem and standardize it in such a way that it is easy to evolve that standard for other things in the future and that will set a model for the way standard attachments might be dealt with in the future as those things evolve and we get a chance to tweak that standard and add to it every year long before it actually gets implemented according to the HIPAA law.
MR. BARR: That is the thing you have to remember is when the other HIPAA regulations were written, they were just updated to meet the current existing HIPAA mandate. They were already in effect. The version may not have been up to date, but all the regulations -- there was a body working on them. They were in the process of updating the regulations. They were working on the process.
This started from ground zero. There was nobody working on this process. There was no X12 body working on automation of attachments. There was no HL7 component working on automation of attachments.
There was just a group of people who started the process from ground zero. There was no transaction to build upon. We had to build the transaction. We have to interface with HL7, which has never been done before. There are interface questions that have to be resolved that are being worked on between HL7 and X12. There are versions. There are update problems that have to be worked on.
There are a multitude of problems that have to be resolved that for the first time came to light when these two organizations met. We are the guinea pigs that are setting the precedent for this process.
The group that comes behind us will have all this ironed out and bless them when they do it. But we have had to make every wrong path and every wrong turn we possibly could to find out the right reasons.
DR. STARFIELD: Okay. A comment from Kepa and then from John and then we will adjourn for the day.
MR. ZUBELDIA: Steven, that may be why I am concerned about something that has not been expressed here yet today. This was started before HIPAA had a mandate with standard transactions. In the list of attachments that you have shown, there are things like explanation of benefits and referrals, workers compensation, things that now have standard transactions.
A concern that I have and I hope I know the answer, but I would like to hear it, is that this could become a black hole. Specifically, today, I have heard something that was very disturbing, is that the workers compensation is different -- the workers compensation transaction, standard transaction, 148, is different from the attachment for workers compensation because the State of New York has a forum that doesn't fit into the 148 and maybe they need to have the forum send us an attachment and that is very disturbing to hear things like that because this could become a black hole that anybody that needs anything that is not in the standard transaction can request as an attachment.
With some of the examples for ambulance that you have in the 48 data elements -- I just saw them on the screen, so I can't tell you which ones they are, but I saw things like date of birth, patient name, patient address, things that, obviously, are in the standard claim. I see no place for them in an attachment if it is an attachment to a claim.
Now, if we are talking about something else, maybe I am missing it, but that is a concern that I have and I would like to have that clarified that this is not going to become a black hole for people that have special needs beyond or different than the standard transactions.
DR. MC DONALD: I just want to respond to that. This list is the literal transliteration of what is on the printed forms. It hasn't yet had expuned(?) those things that are actually -- that are either in the claim -- many of those come back with a 277 and get passed back again with 275. So, these things aren't literal. So, you are right.
The other thing, I agree with you a hundred percent, you have got a problem in not having this be an escape hatch for things that are standardized.
MR. BARR: One of the things we looked at was that we don't look at the 275 as being the escape hatch for anybody who needs a data element. Once we do the data set that we have devised and those data elements that are already on the claim, we are not going to replicate on the attachment because that information is already there on the claim. You don't need to repeat that information.
There is a method, if you need information on a claim that was sent in, that it comes in erroneous, there is a way to resubmit the 837, that is, resubmit that claim data. We are looking about putting data that is not on the 837, that is pertinent to processing the claim.
That data that is on the 837 would not be replicated in the attachment. The black hole question is once we get these items done, we do not want to be the governing body that makes the decision of what should be on a claim or the attachment. We visualize that if you have a request for an attachment or a data element that has to be incorporated in a claim or attachment, you will go to the recognized SDO as a representative and say I need this data element, I need this data element, I need this data element. That SDO would make the call to decide whether it belongs on the claim or it belongs on the attachment. They would refer that request to either the 837 work group or the 275 work group.
We would not be the focal point where people would come to us and say I need this piece of information. There would be a process in place for approval of all subsequent requests.
MR. ZUBELDIA: Thank you. I think that is part of the answer that I wanted to hear. There is another part that says this is not only for the claim but also for all the other HIPAA standard transactions, including workers compensation, including 835, including referrals and some of the other things that are on that list.
Then I have another question, if I may. On the attachment issue, it is not clear to me yet how does this work from the business perspective. If a payer wants to request an attachment with an electronic request, is the provider required to accept the electronic request and act on it or can the provider say I only accept paper requests for attachments?
MR. BARR: The 277 request just for information is not a HIPAA mandate. That is an electronic transaction that was developed in lieu of a proof of concept that was prior.
DR. STARFIELD: I think this is a detail that probably isn't relevant for this discussion.
Go ahead.
DR. LUMPKIN: I guess I just have one concern, looking at the list, and maybe it is because of my years spent in the emergency department, but looking at the list and comparing it to the Proof of Concept Team, I think that I would feel much more comfortable if a rule comes out on emergency department reports and records, if there were someone from the College of Emergency Physicians who was involved with the Proof of Concept Team.
MR. BARR: And we would openly welcome anybody with the expertise that could assist us in defining the questions and what has to be done because the people who are working on this project have no medical or clinical background. We are making a lot of decisions --
DR. MC DONALD: Wait a minute.
MR. BARR: Except for you. We have no one on the team that has a clinical or medical background.
DR. STARFIELD: Bob, did you have a --
MR. DAVIS: I actually thought I was going to be done for the day and my voice is getting a little hoarse.
I agree with Bob a hundred percent as to what is happening with this attachment thing. The responsibility that industry groups, whether it be the Medicaid, you have 50 Medicaid states or have an additional data request, that they would have to get together and decide in some standard way that they have a consolidation. They need to get together.
Kepa, your concern about workers comp and the C4 form in New York State, clearly, the workers comps, there are 50 workers comps groups across the country, they, too, have to get together because unless the industry group gets together to decide what the questions are, then there is no standard. All they are providing for is a vehicle for this to be possible.
The same thing why I am encouraged with the public health people getting together, because we, too, unless we get together and decide on what our common questions are or what our common needs are, there is no standard no matter what the vehicle is. So, I think that this provides a wonderful vehicle to start with, but it doesn't take the responsibility and eliminate that responsibility for industry groups that have to decide their own fate.
DR. STARFIELD: Okay. We are adjourned for the day. We reconvene at 9 o'clock tomorrow morning.
[Whereupon, at 4:50 p.m., the meeting was recessed, to reconvene at 9:00 a.m., the following morning, Tuesday, February 10, 1998.]