[THIS TRANSCRIPT IS UNEDITED]
January 21, 1997
Hubert H. Humphrey Building
200 Independence Avenue, SW
Washington,
D.C.
TABLE OF CONTENTS
Page
Call to Order, Welcome - Dr. Starfield 1
Health Care Identifiers - Mr. Broseker, Mr. Streimer 7
Panel on Health Care Purchasers 86
- Ms. Border, Ms. Koss
Panel on Professional Health Care Providers 131
- Drs. Segal, Rich Milholland, Mr. Jordan
Panel on Health Care Facilities and Other Providers 205
- Mr. Arges, Dr. Schultz, Mr. Rode
BARBARA STARFIELD, MD, CHAIR, Distinguished University Professor/Professor of Health Policy/Pediatrics, School of Hygiene/Public Hlth, Johns Hopkins University, Baltimore, MD
SIMON COHN, MD, MPH, FACP, Clinical Information Systems Coordinator, Kaiser Permanente Medical Care Program, Oakland, CA
KATHRYN COLTIN, MPH, Director, Clinical Measurement Systems/
External Affairs, Harvard Pilgrim Health Care, Brookline, MA
KATHLEEN FRAWLEY, JD, MS, RRA, Vice President, Legislative/ Public Policy Services, American Health Information Mgmt Association, Washington, D.C.
LISA IEZZONI, MD, MS, Associate Professor, Harvard Medical School and Division of General Medicine and Primary Care, Beth Israel Hospital, Boston, Massachusetts
THOMAS LA VEIST, PhD, Associate Professor of Health/Public Policy, Dept of Health Policy and Mgmt, School of Hygiene
& Public Health, The Johns Hopkins University, Baltimore, MD
JOHN LUMPKIN, MD, MPH, Chair, Work Group on K2 Health Data Standards, Director, Illinois Department of Public Health, Springfield, IL
VINCENT MOR, PhD, Director, Center for Gerontology and Health Care Research, Brown University, Providence, RI
CLEMENT JOSEPH MC DONALD, MD, Co-Director, Regenstrief Institute, Indianapolis, IN
GEORGE H. VAN AMBURG, State Registrar and Chief, Office of the State Registrar and Center for Health Statistics, Michigan Department of Public Health, Lansing, Michigan
WILLIAM BRAITHWAITE, MD, PhD, Senior Advisor, Health Information Policy, Office of the Assistant Secretary for Planning and Evaluation, DHHS, Washington, D.C.
MARJORIE GREENBERG, Chief, Data Policy/Standards Staff, Office of Data Standards, Program Development and Extramural Programs, NCHS, CDC, Hyattsville, MD
ROBERT MOORE, Director, Office of Systems Management, Health Care Financing Administration, Baltimore, MD
P R O C E E D I N G S (9:05 a.m.)
DR. STARFIELD: My name is Barbara Starfield and I am chair of the Subcommittee on Health Data Needs, Standards and Security. I am hereby convening the meeting of the subcommittee and of this special work group on administrative and clinical data, fondly called E-2.
We will not dilly dally over the charge of the subcommittee this morning and delve directly into business.
First, I want to do a couple of things, the first of which is some introductions. So, we will start around the room with Bob and go this way and then wend our way around the back.
MR. GELLMAN: I am Bob Gellman. I am a privacy and information policy consultant.
MR. SCANLON: I am Jim Scanlon with HHS. I am the executive staff director for the national committee.
MS. GREENBERG: I am Marjorie Greenberg from the National Center for Health Statistics and acting executive secretary of the national committee.
DR. COHN: I am Simon Cohn. I am the clinical information system coordinator for Kaiser Permanente.
MR. MOORE: I am Bob Moore. I am the HCFA liaison to the committee.
DR. LUMPKIN: John Lumpkin. I am director of the
Illinois Department of Public Health and a member of the committee.
DR. STARFIELD: I am Barbara Starfield. I am professor of health policy at Johns Hopkins University and chair of the subcommittee.
DR. BRAITHWAITE: I am Bill Braithwaite, senior advisor on health information policy at HHS.
DR. MOORE: I am Vince Moore. I am a member of the committee and chairman of the department of community health at Brown University.
MR. BROSEKER: I am Joe Broseker, director of the Office of Health Care Information Systems with the Health Care Financing Administration.
MR. STREIMER: I am Stewart Streimer. I am director of the Office of Program Requirements in the Health Care Financing Administration.
MR. MAYES: Bob Mayes, Health Care Financing Administration.
DR. MILHOLLAND: Kathy Milholland(?), senior policy fellow with the American Nurses Association.
MS. EMERSON: Mary Emerson, Health Care Financing Administration.
MS. TRUDEAU: Karen Trudeau(?), Health Care Financing Administration.
(Further introductions made off microphone.)
DR. STARFIELD: Thanks very much. As you know, the Administrative Simplification Provisions of the Kennedy-Kassenbaum legislation give the National Committee on Vital and Health Statistics a lot of responsibilities in a relatively short time.
So, we will devote these two days primarily to our K2 working group, and I in a moment will turn the chair over to Dr. John Lumpkin, the chair of that working group.
First, I would like to just review the agenda for these two days. We will start off with presenters who will talk about health care identifiers.
Following that we will have a break and then we will have a panel with people who are representatives of health care purchasers.
After lunch we will turn to a panel of professional health care providers, after the next break a panel on health care facilities and other providers. Then we will move into some general subcommittee business.
Tomorrow we will convene probably at 9:00 o'clock but, depending on how today goes, perhaps a little earlier, then move into some representatives from health plans, who will speak on the administrative simplification provisions, then to a panel of payers, and then move to our subcommittee business in the afternoon, probably heading toward adjournment about 4:30.
I would like to ask the subcommittee members if they have any comments about the draft minutes, which you should have received in the mail or the e mail in the last couple of days. We need your final comments by January 27th, but are there any things we need to be aware of at this time?
Then without further ado, let me turn the floor over to John Lumpkin.
DR. LUMPKIN: Good morning. I would like to welcome everyone and thank you for coming out today, braving the post-inaugural Washington. Certainly it was something that many of us felt to be a challenge, especially some of us from smaller cities like Chicago, where we don't go through this kind of thing except on St. Patrick's Day.
We have really a fairly formidable challenge in addressing the administrative simplification issues. I would like to sort of start with you with a little bit of my bias on where I hope we go and some of the questions and issues that I hope you will raise.
My bias really starts out that my career began in the clinical practice of emergency medicine. Given that time, which has been a few years in the past, I can still remember the frustration of a practicing clinician who spends more time buffing the chart and dealing with the paperwork, than treating the patient and coming up with clinical challenges.
Even more pertinent than that, now, are some of the experiences, because my wife is a practicing clinician. I know, as I go into her office over the last few years, the numbers of forms and papers that are sitting in her office have increased astronomically.
One of the ventures of having multiple managed care organizations is that each one uses a different lab and each lab, of course, has its own requisition forms and accompanying forms and clinical forms and billing forms.
Now that she has to deal with three different labs, the amount of space that is taken up on her desk has increased significantly.
Then when you add in all the other forms, the requests for approval for surgery and so forth, we can understand that the clinician of today bears little resemblance to some of the clinicians of the past.
About 10 years ago I did some consulting work with HRSA in Egypt with the Ministry of Health. I visited a little hospital in a small town right outside Luxor.
This little hospital had this emergency department which I was reviewing. Inside this emergency department they were telling me -- it was a very small room, it is smaller than the size of these tables -- they said they saw 25,000 patients.
I was really curious as to how they could see 25,000 patients in a room that size. So, I said, could I see your charts.
They looked at me and said, charts? If we had charts, we couldn't see that many patients.
So, we really hopefully, if we address the issues of administrative simplification, we can also have a dramatic impact on the treatment and practice of taking care of those who are ill.
Frequently we are in the position -- and these are the kinds of questions that we are going to have to ask -- how simple is simple enough. How many variations of a form are really necessary.
We really need to, and want to, throw that out, because these are some very important questions and there will have to be a balance struck, obviously, the need for variability, the different requirements, the different plans.
We are trying to look for a reasonable one, and I think we have passed beyond a point of reasonableness, or at least that would be my interpretation of why Congress passed this section of Kennedy-Kassenbaum.
Any comments from the committee before we get started?
Then the first item is the health care identifiers.
DR. LUMPKIN: One of the things that I would like to make everyone aware of, as you recall in the legislation that was passed, unique identifiers are one of the areas that standards have to be developed for in this legislation.
HCFA has been working on two of those identifiers for some time now. One is the payer identifier and the provider identifier.
In our initial pass through what we have gotten from the standard organizations, we believe that HCFA has on the table two of the items that we would like to propose to the Secretary as being what she would adopt as a national standard.
What you are going to get this morning is a presentation from Stewart Streimer, who is working with the payer ID, and Joe Broseker, who is working with the provider ID.
One of the things that we are required to do in the legislation is also present to the committee where we are going and where we are and what we intend to do, as the standard group works toward developing these standards that we are going to recommend to the Secretary for adoption.
We had made a presentation more than a year ago to this committee on the payer and provider ID. What we are back to do today is to represent, because there are some new members on the committee, also the legislation has passed.
So, we are going to go over this again and we fully intend to put this out. On the provider ID, I think Joe will tell you where we are with the regulation and where we are coming from.
MR. BROSEKER: I would like to thank the committee for giving us the opportunity to come here today to present information on these identifiers. And a special thanks for putting me on the agenda ahead of Stewart Streimer, because Stewart always brings color copies of the handout and he is a hard act to follow. So, I am happy to be leading off.
My presentation is going to cover the national provider identifier. We say it is a proposed standard provider identifier. We mean exactly that.
We are proposing it. We are expecting to publish, or hoping to publish, a notice of proposed rule making in February, on or around February 21 of this year, proposing the national provider identifier as the standard identifier for providers.
We will also be announcing how we propose collecting the information and in what categories or what process we will begin enumerating providers, how we will do that initially, as well as what kind of data we will be collecting, what kind of data we will be releasing.
Some parts of that MPRM are certainly not final. One thought behind getting the rule published as quickly as possible is that we still know, in spite of all the consultations we have done with the industry, there are still some segments of the industry that perhaps are not as well aware of what is happening with administrative simplification as they need to be.
We want to get the rule on the streets so that we can get feedback from the industry, consider that carefully, and learn from it before we publish the final regulation, which we hope to do in July of this year.
We then plan to begin enumerating Medicare providers and issuing their national provider identifiers in very early August of this year, and then require the use of this identifier for Medicare claims processing starting December 1st of this year.
So, it is a rather ambitious schedule. Considering all the standards that have to be adopted by the secretary under the administrative simplification provisions of the legislation, it is imperative that we get a very quick start and see how this first reg goes.
We are sure we will have to make a lot of changes. We have to learn a lot yet. We want to learn that as fast as possible, so that we can build that learning experience into the other regulations that have to follow.
On my next slide -- I will try not to do this too much during the presentation, but being a systems person, I can't help but use acronyms.
I will just show you the list of acronyms that I will most frequently use. NPI refers to the National Provider Identifier. NPF is the national provider file that will contain the baseline information identifying and enumerating providers.
NPS is the national provider system. That will be the system designed to house the information identifying providers and their numbers, as well as give us the edits to make sure that the data that we collect in that system are as accurate as possible and are as comprehensive as possible, so that we can avoid issuing duplicates, or issuing more than one number to one provider.
I will especially try to avoid using that acronym of NPS because it sounds too similar to MTS, the Medicare transaction system. I will talk a little bit more about the Medicare transaction system in a later slide.
UPIN is the unique physician identification number that is currently used for Medicare physicians. Then PIN is provider identification number, which is yet another number that is normally issued by Medicare contractors to Medicare providers. I am sure that term is probably used by other payers as well to identify their number.
Today Medicare providers have multiple PINs. Certainly that is one of the things that we are trying to avoid with the NPI.
Just briefly a little bit about how we started, MTS, the Medicare transaction system, is a major HCFA initiative that has been underway for several years.
The purpose of that system is to consolidate a number of different standard processing system, if you can have a number of different standard systems. That is certainly a contradictory terms.
It is trying to consolidate those systems into one standard system that will be used nationwide for processing Medicare claims.
As a forerunner to putting that system up, one of the things that a standard system will require is a standard system of identifying providers who will submit claims.
Today, as I mentioned, Medicare providers can have any number of PINs to submit claims, and the PIN is what is used in Medicare billing processes.
As a result of MTS, we realize that, at least for Medicare purposes, we had to have a standard identifier for Medicare.
At the same time that we were looking for that, there was legislation pending on the Hill -- at that time it was the Health Security Act -- as well as a number of other legislative actions, many of which included provisions on administrative simplification.
Knowing that that pending legislation was out there, we decided that it would be a fatal mistake for us to look at just Medicare, and that we ought to look at identifier and a system that could support just about any payer and any group of providers in the country.
Even though we didn't have legislative authority to go beyond Medicare and Medicaid at the time, we at least began talking to entities outside those two programs, in order to construct something that would suffice under administrative simplification, if and when it passed.
I guess it is fortunate that we did that, because the legislation, as you all know, did pass last August.
One of the things that we certainly wanted to avoid was the expense of making modifications to Medicare programs, the burden of getting Medicare providers to change to a number and then, because it wouldn't suffice to satisfy other programs, to have to move to something else under administrative simplification.
So, we have tried to keep the national community in mind from day one.
We have also been involved with standards activities with a wide variety of groups; ANSI, the work group for electronic interchange, WETI, the national council for prescription drug programs, a number of groups we have had regular contact with and discussions in terms of getting their input into what we are doing and making sure they are aware of what we are doing.
The next slide talks about some of the groups that we have dealt with, just to give you a high level overhead. As I said, we certainly have not done this in a vacuum.
We have dealt with standards organizations. We have dealt with Medicaid programs, state agencies. Several state representatives have been part of the design and development team from day one.
We have talked to a number of professional associations, other government payers including CHAMPUS, the Department of Defense, the Office of Personnel Management, Social Security Administration, Food and Drug Administration, just about every federal payer of health benefit plans.
We have also had a lot of discussions with groups in the private sector and provider groups. We have talked to the Practicing Physicians Advisory Council, made a presentation to them.
We have talked to the American Dental Association, the American Medical Association, numerous organizations, as well as folks in the private sectors, whether they be payers or organizations devoted to processing bills, clearinghouses, for example.
So, we have had a lot of discussions with a lot of organizations and have learned an awful lot from them in constructing the design of the NPI.
What is the national provider system, and what is it not? It is important that we be able to distinguish those because some folks expect the national provider system and national provider identifier to do a lot of things that PINs and UPINs have been doing over the years.
As a result of PINs and UPINs being used to do things other than just enumerate and identify, folks are expecting an awful lot of the national provider identifier.
Basically, we believe that the national provider system is simply a system to enumerate, to collect enough information to make sure that we enumerate providers only once, and that a provider who is sanctioned, for example, can't go to another state and try to re-enroll and get another number.
It is designed to just have basic demographic information so that we know who the providers are. It is not designed as an enrollment system.
By that we mean, every payer in the country may have different standards, different conditions of participation, for providers to be able to enroll and bill for services to that payer.
So, the national provider system is not an enrollment system, per se. It won't tell you whether or not a provider who has an NPI is eligible to bill the Federal Employee Health Benefit Program, for example, or is eligible to bill for CHAMPUS services. Those kinds of decisions are left to the payers.
We believe with the country having thousands of payers, it would get too unwieldy to try to have one system to try to keep track of all the information that determines whether or not a provider is eligible to bill this program or that program.
So, we are not trying to be an enrollment system; nor are we trying to be a claims payment system. We are not trying to collect information that allows payers to price claims, for example.
Again, the algorithms for doing that are different among payers. To try to include all those differences in one system would just be too unwieldy.
We believe that the national provider system, just for the function of enumerating and identifying providers is going to be large enough and complex enough without including pricing information, for example.
When we talk about administrative simplification, standard identifiers and basically any standards are good, but there are always caveats.
A standard will not always be exactly what everyone would like to have. That is probably going to be the biggest challenge in implementing the standards under administrative simplification.
The fact that you are going to have a standard -- if you have a standard that pretty much pleases everybody, then basically you have probably developed a standard that is so complex and so large and so all encompassing that you end up pleasing almost no one. You don't end up achieving administrative simplification.
So, we point this out because any standard is probably going to require some folks to change. It is going to cause extra work for some.
Folks who are using PINs, for example, to identify locations or identify specialties of a provider are not going to be able to do that with the NPI. So, they are going to have to make some changes.
I can tell you that HCFA has been very keenly aware of this because our own systems are going to be changing drastically as a result of the structure of the NPI. I will get into that structure in just a few minutes.
We did not pick something that looked like just an expansion of the UPIN, for example. The UPIN was not going to work for HCFA in the not-too-distant future. It is limited because of its size and the way it is constructed.
It is not going to be the same as PINs that we use today in the Medicare program. So, by getting away from those two standards we caused an awful lot of work for our own internal systems.
What we did do was come up with something that we believe can be embraced, and has been embraced, by many in the outside world.
It has been endorsed by the American Dental Association already, the National Council for Prescription Drug Programs, I believe NUBC and the Massachusetts Health Data Consortium, the state of Minnesota. A number of other groups have embraced and endorsed the concept and the construct of the NPI.
Exactly what is that construct and format? The NPI is designed as an eight-digit number. It will be alpha numeric, although for the beginning of the NPI it will only contain numerics.
That is because there are a number of systems out there in the Medicaid arena and in the private pay arena that can only handle numeric identifiers today.
We suspect that all those systems will be undergoing changes to accommodate the other standards under administrative simplification.
In order to give those systems more time to do that, we decided that initially NPIs would only be issued with numerics, although eventually we will assign alpha numerics.
By that time, all the other standards will have been adopted and implemented.
The NPI is eight positions. Seven positions are unique that identify the provider. The eighth digit is simply a check digit that can be computed by doing different types of manipulations with the first seven digits.
Those first seven digits are manipulated using an algorithm that is in the public domain and we are going to be publishing that on the HCFA web site, so that everyone can see how we are computing the check digit.
Basically, the check digit helps eliminate keying errors, especially for systems that start off using pads such as a telephone pad to key in numbers for claims processing.
There is also something called a two-position location code that identifies for Medicare purposes, at least at the outside, the location of a provider.
A physician will only have one NPI, but they may practice in multiple locations, and their claims may be paid at different rates for different locations.
I will point out quickly that location code is probably going to be one of the more controversial aspects of developing a provider identifier.
In fact, we are now looking at the location code as a separate data element, because we are not sure what the industry is going to tell us about a location code.
We do know that there are a number of folks who are saying, why not just use a five-digit zip code plus four, and let that be the location code.
For many payers, that will work. For some payers, that may not work. Some payers -- a provider, for example, could have an office in one location and the nine-digit zip code would suffice in identifying that office.
But the provider may be working for two different plans, two different managed care organizations, who are going to price the claims differently, depending on what time of the day or what floor of the location the provider was in when they were practicing.
So, will a nine-digit location code work based on the nine-digit zip work? We don't know for sure yet.
Will folks have to change and somehow hone in on nine digits and make that suffice for pricing claims? Again, we don't know. We would like the industry to give us input on that as a result of publishing the NPRN.
I would also say that we are thinking these days that that really is a claims processing issue, not necessarily a provider identifier issue. We would look to the group that develops the standard claims format for addressing that issue.
In either event, the national provider system will eventually house either a two-digit code, a nine-digit zip code, which we are planning to house anyway, or some other location code that the industry develops pursuant to the claims processing standard.
So, we would like to see the location code kind of divorced from the provider identifier as an issue.
The last item on this sheet also is probably a controversial one. It is one that has had a lot of impact and will have a lot of impact on Medicare claims processing.
That is, we are building no intelligence in the NPI. That was a requirement that was laid down by the standards group, especially the HISPP, the health informatics standards planning panel, which I believe has been renamed HISB.
One of the requirements for getting the intelligence out of an identifier, that is one of the requirements those groups imposed on anybody developing a standard identifier for provider.
Basically what this means is that you cannot look at the NPI and know anything about that provider, in terms of what state they are in, what their specialty is.
For example, some Medicare PINs today, the first two digits tell you what state the provider is in. You are not going to be able to tell that from the NPI.
One of the key reasons for that is, as soon as you introduce intelligence into an identifier, that means whenever any activity related to that intelligence changes, you now have to issue another identifier.
So, you can't then have a unique standard identifier. If a provider is working in two states under today's system under Medicare, for example, they have two PINs.
That gets you past the requirement to have a unique identifier for providers, no matter where they practice in the country.
So, we have eliminated intelligence from the NPI. Again, there are a lot of systems that have been counting on that intelligence to tell them things, especially in the audit and reimbursement area.
Those kinds of systems will have to change to key off whatever other data elements they need to identify that intelligence.
The types of providers that we will be enumerating in the national provider system are individual providers, obviously. We will also be enumerating groups, which we consider two or more individual practitioners.
Of course, organizations, institutional providers, and we will also be able to link group members to a group. We will be able to know the NPIs of individuals and relate them to the NPI to any group that they practice in.
The type of data that we will have in the national provider file includes demographic data and the information that is actually in the national provider file, the data elements, we have already published on the HCFA web site.
So, folks wanting more detailed information on exactly what data elements we will have for what types of providers is available via the internet.
We will have location data, sanction data starting off with HHS sanction data, but certainly to help implement other pieces of legislation that have been passed over the last couple of years.
We will also be looking to gather sanction data from other programs as well, and we will also have other identifiers that relate to an NPI.
There are SSNs for individual practitioners, if they choose to give us their SSNs, although at the moment they are not mandated to give us their Social Security numbers.
We will also have employee identification numbers, tax ID numbers, UPINs and any PINs that are associated with NPIs.
When we convert and load providers into the file initially, any numbers that we are aware of from other sources we will also include in the file at that time.
That will be especially helpful in giving folks cross-walk files during a period of time when we are first starting up.
People who do have PINs and UPINs will be able to use a cross walk file to find out what the NPI is.
The features of the national provider system will be supplying basic data validation. We will be looking at Social Security Number, other types of data elements to try to ensure that we don't enumerate individuals more than once, and to ensure that we do, in fact, have valid providers coming into the system.
As I mentioned, we will be matching against the sanction file from the Office of the Inspector General of HHS initially.
We will be doing searches for duplicate providers, preventing providers from getting more than one number. We will also have query and report generation capabilities from the national provider system.
We will be publishing something similar to the UPIN directory today, although it will probably be more comprehensive to include more information about providers than you find in the UPIN directory today.
That information is probably going to include address information, not just names and zip codes and the UPINs.
That is one area also that we are looking for public feedback in response to the NPRM, what kind of information should we release, what kind of information about providers should we protect.
The enumeration process, as we plan to do it right now, initially we are going to load Medicare providers, and we will take existing files, whether they be UPIN, the National Supplier Clearinghouse, or OSCAR -- OSCAR stands for On-line Survey Certification and Reporting System, and that is the system that we use to certify institutional providers.
Any provider currently having a number under any of those systems will automatically get a national provider identifier when we do the initial load, and we will be starting to do that in a few months.
Then we will, as new providers come into Medicare through the Medicare enrollment process, they will receive their NPIs and we will stop issuing UPINs and other numbers as soon as we bring the NPI up live.
Non-Medicare is to be determined yet. We do know of at least one state that wants to begin using the NPI at the exact same time as Medicare does.
That state is Utah, because they have a rather extensive network in Utah already for processing physician claims and they don't want to disrupt that.
At least their initial thinking is that they might want to institute a requirement to submit the NPI for all claims in Utah at the same time as Medicare does.
Exactly when we will bring other payers in - Medicaid, FEEB, CHAMPUS -- is yet to be determined.
One thing we would like to have is, I will call it, a little bit of breathing room. With any new system, there are going to be some start-up problems.
We are conducting extensive testing of the system, not only with Medicare but with agents who deal with Medicare.
We are soliciting volunteers from other sectors of the health care arena to participate in some baseline testing of the NPI, to try to minimize problems.
As much as we would like to say there won't be any, I am sure there will be some. We certainly hope that they will be minimal.
To the extent that there are problems, we would like to get through those with the Medicare program first, so that we get those problems out of the way before other entities come in and use the NPI.
Just some final thoughts to remember. One size fits all, but it is like those T-shirts at the resorts where one size fits all.
Some of us, it will fit quite comfortably and some it won't. But the benefits are enormous to all the stakeholders, and especially to providers in terms of simplifying life for them a great deal.
The benefits will certainly outweigh the changes that have to be made initially by those to accommodate the NPI.
There are going to be some operational and implementation problems. We are committed to working with the industry to minimize those and to try to work them out so that no one is unnecessarily over-burdened by the changes.
Status information on the national provider identifier and system are on HCFA's web site, www.hcfa.gov. There is also an e mail box there where anyone can submit questions and the national provider system team will get back and respond to those questions.
We are certainly hopeful that this committee will also endorse the NPI. At this time I would be happy to take any questions that you may have.
DR. BRAITHWAITE: Joe, clearly this identifier has to do with more than just physician providers. What is the range of providers that is being considered here?
MR. BROSEKER: Our guesstimate is that eventually as many as three million providers will be enumerated using this system.
We have been working with NCX12 to develop a taxonomy that includes the range of providers that may eventually have to be enumerated by the system.
We have constructed the identifier, in its eight-digit size, to accommodate literally -- I think the number is over a billion providers eventually.
So, we are starting with what we think is a maximum at the moment of three million. We believe that the construct of the numbers is going to last us well into the next century, if not beyond.
Right now HCFA has approximately 800,000 providers -- physicians -- enumerated in its UPIN process. We have over 100,000 clinical labs, and I think somewhere in the neighborhood of 15,000 long-term care facilities, another 6,000 or 7,000 hospitals, as well as over 100,000 suppliers.
So, HCFA currently has enumerated approximately 1.2 million providers in its existing UPIN, NSC and OSCAR files.
We will have to add to that eventually approximately 200,000 dentists. The ADA has information on better than 90 percent of those.
We will have to look at the providers that HCFA does not normally enumerate through the Medicare program -- obstetricians, for example, pediatricians, psychologists, psychiatrists.
There is some question as to whether or not RNs should be enumerated. I think at the moment the providers that we are concentrating up front are those who are involved in submitting claims and actually being reimbursed by payers for services.
As you look down the road to computerizing patient records and clinical information, we think it is going to be important also to be able to enumerate RNs, those who actually give services but may not separately bill, but so that you have that information in the patient record.
So, we expect to eventually enumerate several million. That taxonomy, I believe, if you all are not aware of it, we can certainly make that available to you. It is important to know that the NCX12 is the one who is actually going to manage that taxonomy for us.
DR. MOR: One of the types of providers that HCFA now has for Medicare is the home health agencies. Each of those home health agencies -- that file has been growing substantially over the years -- has different sets of numbers if you go into the Medicaid world in different states.
Then within that Medicaid world of the home health agency they have subcontractors that provide, under bonded or non-bonded services, of all different sorts.
There are a number of demonstrations now underway in states across the country to have these bonded agencies providing personal care assistance services to various classes of Medicaid recipients.
Any thought about whether those should be individual providers, or because they will be billing the Medicaid system under waivered or non-waivered systems as individuals, or would that be under the organizational thing? Sort of, that is the most distant.
MR. BROSEKER: It will be sort of a combination of things. We will see them as individual providers in the sense that if they are billing Medicaid in several states and they are also billing Medicare, they will only have one number to do that.
If they are in separate physical locations that they bill from, where their principal office is but they are also tied to headquarters, they are tied to a chain, individual pieces of that chain may have their own numbers, will have their own numbers.
We are also looking at developing information that will allow us to link them back to the chain. So, in some ways they will look like individuals. Again, they will only have one number per location and we do hope to link them back to the chain.
DR. COHN: Thank you very much for your very informative presentation. I do think the committee would love to see a copy of the taxonomy that is proposed by X12. I think it opens up what you means by an individual.
Certainly, as I began to think about it, I assume that means pharmacies and it just goes on and on. One could almost imagine that managed care organizations could be considered providers, except maybe they are also payers. So, one might wonder where that edge is and I think it might be very useful.
Now I actually just had very specific questions about this location code and how you would impose that as well as trying to make decisions on it.
I guess I am a little confused because you talked about having proposed rules coming out in February. Yet, it seems that there are significant questions that you have around location code at this point.
Are you at this point completely separating the two and having them on separate tracks?
MR. BROSEKER: Yes, we are separating the two, in terms of not being related to the provider identifier. We believe that the regulation on the provider identifier can go and can be implemented.
Using the two digit code as a separate data element in the system, that will suffice for Medicare claims processing purposes.
If the industry says, now we want a nine-digit zip code to use for location code, we will also have that in the system.
If the industry says, as a result of the processing or developing the reg for claims processing, that the processing code becomes the nine-digit zip code, we will simply begin using that in Medicare processing.
What happens if the industry says there has to be something above and beyond that, we don't know yet. We have had some folks come in and say we should actually enumerate physical locations.
There are other payers who have said, that won't quite work because, you know, some folks would like a physical location to always have one unique number and always key off of it.
Number one, the legislation doesn't authorize us to enumerate locations. Secondly, with the example I mentioned earlier where the provider is at the same location, but at different times of the day may be providing services to patients with different plans, even uniquely identifying that location wouldn't work.
When that was brought up, some other payer said, well, we also need to know the time of day that the provider gave the services at that location.
We didn't quite see that as administrative simplification. So, we said, some folks are going to have to change. Also, location code, then, is not something that has to be answered now, and let those who are going to develop a claim standard address that issue.
It is more related to processing and paying a claim, than it is to identifying a unique provider.
DR. COHN: Just to make sure that I understand, you are separating the two standards, but you are continuing to implement the location code, standard, ending with something else.
MR. BROSEKER: That is correct.
DR. COHN: So, it is not really being separated, then.
MR. BROSEKER: It is separated in the sense that Medicare is going to go ahead with the location code for its purposes.
Hopefully by the time we implement it, we will have a clearer sense of what the claims folks are developing, and whether or not location code is, in fact, an issue.
It may be that some payers are going to have to make some changes and not hone in on a specific location and time of day, for example, as to when services were rendered, as to how they will price the claim. Maybe they work off something else.
MR. MOORE: I think one thing here -- when I hear some of Simon's questions -- is that I think when we started out we got wedded to a situation where people had, over time in their processing, identified location and provider identifier as one.
As we got further into this process we began to realize, when looking at different processes around the country, that we were uniquely identifying the provider, not necessarily where he was.
That was something else that we had to address, but making that part of the identifier of the provider, the location of the provider.
Utah was a great example where they said they had 1,200 locations for a single provider. We thought that was impossible, but that is what they allege.
We didn't know what else we were going to encounter that had to be part of whatever the vehicle was, whether it is a claim or whatever.
There is a data base that supports this unique identifier. The unique identifier is just one element in that data base.
The other information about that provider -- location -- could be added to that data base, just like you would any other data base of information. It just depends on what you want to use and how you want to use it. We are struggling with that.
One of the things that I want to mention, one of the reasons we put out the reg is not just to tell everybody, this is what we are going to do.
We want to tell you what we propose to do, so that you can come back to us and tell us why you think we might be taking the wrong approach in where we are going.
This is not a one-way street. This is something that I stressed to OMB, when I talked with OMB back before the holidays.
They think they are going to have a very big role -- don't quote me on that -- in this standards process. But these are national standards applied to all the health care community.
We don't think that we know all the answers. In fact, we know very few of the answers, because we have seen the program from the Medicare/Medicaid point of view for a long time.
As we get out there into the commercial world, we are going to see, I think, a whole different way that they have been doing business. We have got to support that, we believe.
So, we are going to be as flexible as we can coming out with these regs, telling people as much as we can about where we are coming from, what we are doing, and we may even be asking questions in the reg to say, you know, this is a problem that we are wrestling with.
All of those who have wrestled with this before and solved it, please let us know of the solution so that we can factor them into our thinking.
DR. STARFIELD: Could you help us clarify what you mean by provider? We understand it is not the same thing as the biller, and we presume there will be another identifier for billers. Is that right? Is a provider a biller?
MR. BROSEKER: A provider could be a biller.
DR. STARFIELD: Could be, but is it necessarily?
MR. BROSEKER: Could be, and initially the providers will be enumerating our billers. We may be enumerating other providers eventually who are not billers.
Right now the term that we are going with to define providers is basically that which is in the legislation that was passed.
In essence, it boils down to anyone who provides health care services to an individual is considered a provider.
DR. STARFIELD: A lot of people provide services and don't bill for them.
MR. BROSEKER: That is correct.
DR. STARFIELD: How do you get around that?
MR. BROSEKER: I think that the taxonomy that NCX12 has developed would help identify that. For example, you might see a dental hygienist, RNs, dieticians, some folks who are not necessarily billers in today's world, at least in today's Medicare world.
They do provide services and eventually they may need to be enumerated.
DR. STARFIELD: I think pretty soon we are going to have a situation where most providers are not billing for their services.
How are you going to deal with that? Are you going to have separate numbers for the billers and separate numbers for the providers?
MR. BROSEKER: Not at this point. The only time that I guess in terms of managed care organizations for Medicare purposes, they could be billers. But the managed care organization could also be a provider.
Our intent is to issue numbers to managed care organizations as if they are providers. They will also, as you will be hearing, receive payer identifications as well. We provide managed care organizations as providers and payers.
MR. STREIMER: Dr. Starfield, does your question go a little to the fact that we are going to have more and more billing agencies and clearinghouses, et cetera, coming into the marketplace, doing the billing on behalf of providers?
DR. STARFIELD: My question is really directed at, what kind of information are you trying to get by getting a provider number?
Are you trying to identify clinical services to a provider of those services, or are you just interested in billing?
MR. STREIMER: I think the answer is yes. I think the premise is that we would want to identify the providers of health care associated with the rendering of that services, the providing of that health care.
If that health care is billed to a particular payer, we would want to know what provider is associated with that billing.
I think as Bob and Joe had pointed out earlier, we are not just talking about the billing environment in terms of claims.
We are looking at all the possible transactions that deal in the health care industry, and where a provider identifier might relate in each one of those particular transactions.
As Joe indicated, if you are within a hospital setting and you are a nurse who wants to share information with another part of the hospital, there is really no billing taking place, but to identify that particular practitioner on that transaction, we would want to uniquely identify that practitioner.
DR. STARFIELD: That is fine. I really would urge you, that in this tremendously valuable process you are going through, to think about what you mean by a provider and why you need to know what it is, because the world is changing.
DR. BRAITHWAITE: Particularly since it sounds like you are identifying all billers as providers, even though they might not actually be providing care.
DR. LUMPKIN: Let me not let you loose on this particular interest, because I heard the phrase, in defining provider in today's Medicare environment.
I think our charge as a committee is not to look at an identification number for Medicare, but for all entities that will be involved in the health care system and across all providers.
I just did a couple of simple, a bit of mental arithmetic, based on what we do in Illinois. We have about one-fifteenth of all nursing homes in the country.
We as a state are required to register every nurse aide under federal law. They are providers of care. Are we going to have a separate enumeration process for those certified nurse assistants, who are required to be on a registry in every state of the country, and then a separate system for other billing-type providers of care?
If that is the case, I suspect that the number 10 million, which is what I get from a numeric only eight-digit number, will not be adequate, particularly if there is any turnover in the short period of time, until you move to the alpha system.
So if, in fact, we are defining provider as someone who delivers health care services and we have not scoped out the universe adequately enough, who makes a determination of what entities will fit into that system, and what is HCFA's commitment to provide that service, not just to Medicare/Medicaid, but to the entire health care delivery system.
MR. BROSEKER: Our commitment is to support the entire health care delivery system. We believe that the NCX12 document goes a long way to covering just about any type of provider, going from individual practitioner to taxi drivers who are providing ambulance services. That is how wide ranging the NCX12 taxonomy is.
In terms of the 10 million limitation, we would certainly hope that once a standard is developed, not just this standard, but a standard for all the transactions that are required under the administrative simplification legislation, there are a number of transactions where the standards have to be developed within 18 months of passage, or February of 1998.
Then they have to be implemented by everybody transmitting health care data electronically by two years after that.
So, we don't see that the time frame for using alpha numerics is that far down the road. It is no more than February of the year 2000, where everybody would have been required to be using those standards.
Whether or not we could enumerate more than 10 million in that period of time, I am not sure. I don't think so.
In what we have seen, in terms of just developing and converting the Medicare files and bringing Medicaid in, that is going to be quite an undertaking.
Whether or not the industry tells us to bring in all those other groups on day one or by February of 2000, or whether the rest of the industry is going to struggle just to make sure they implement all the standards by February of 2000 and not enumerate other groups of providers they haven't enumerated up until now, I think that is going to be a significant challenge.
Whether or not we would run out of the 10 million before those standards are fully implemented, I am not sure that would happen. It certainly doesn't seem likely.
DR. LUMPKIN: The way you are describing, even though you say there is no intelligence involved in the code, if you are going to phase in providers, you are going to have to reserve, at least in the initially phases -- for instance, if 2001 is when you decide to merge in the nurse aides, obviously by going to an alpha code at that particular time, there will be some encoding, at least in the early phases, where there will be some numeric ranges based upon phase-in of different provider groups.
MR. BROSEKER: I am not sure I --
DR. LUMPKIN: Let's assume you have 10 million providers in the year 2000. January 2000 you decide to add in now nurse aides. They are going to be the first alpha grouping.
If you bring in seven million nurse aides, then the next seven million numbers that have been rolled in will all be nurse aides.
Then the next group that you bring in, which may be personal assistants, that will be the next group that will be phased in, they will all have the next series of alpha numbers.
MR. BROSEKER: The numbers are going to be assigned using randomizing techniques. The first number assigned could start with a digit of eight, for example.
DR. COHN: There is so much here to ask, I don't know where to stop.
I guess I am wondering, first of all, whether the proposed regulations, part of that will be identifying what your initial scope of providers are. Is that part of the proposed regulations?
MR. BROSEKER: In terms of identifying how we will initially load the data base? That is in there. Medicare providers.
Then we are asking the industry to comment on how to bring other types of providers in, and how many entities, for example, should be doing that.
Should every payer have the authority to identify and, via the national provider system, cause numbers to be issued to their providers?
Should we have just state licensing boards do it or maybe state Medicaid agencies?
One of the things we hope is that there won't be many entities per state who will actually be interfacing with the national provider system to collect and submit baseline information.
We hope that every payer in the country doesn't have to do that, because that would be too many entry points into the system. It would require too much work, not only on payers parts but also on the government's part.
So, one of the things we are hoping for is that the industry will propose some ideas as to how providers get enumerated beyond the Medicare program.
Again, one entity per state, maybe the professional associations, should do it, maybe the provider boards. That is an open question yet.
DR. LUMPKIN: Let me follow up on that question, because certainly I think, as I understand it, this will come before the full committee in terms of recommendation to HHS in regards to this particular component of administrative simplification.
If, in fact, one of your suggestions, which is to use state licensing boards, it raises the issue of the process of adjudication of a similar type request for identification.
If you roll over the -- let's say for example you roll over all the Medicare providers. Then in the next phase you go to all the state licensing boards and you roll over all the nurses and physicians.
You very likely will have a very high degree of overlap. What kind of mechanism and how labor intensive is that mechanism that you envision of adjudication of those kinds of providers who might be the same individual provider, even though some of the data elements may not be exactly the same?
MR. BROSEKER: Because they practice multiple specialties, for example?
DR. LUMPKIN: No, the state licensing board and yours might have a different address. Those are, again, some kind of minor differences, but they do require at least that there is some adjudication that this is, yes, the same person or not the same person.
MR. BROSEKER: We are going to be looking at the same basic demographic information initially to try to weed those duplicates out, things such as name, date of birth, sex, hopefully Social Security number or tax identification numbers.
We also have software that will compare addresses, if there are minor differences in the address. They will be detected and we will know they are the same address.
That would be a secondary match criteria. We would key off primarily name, date of birth, sex, Social Security number and other identifiers that we are aware of.
DR. LUMPKIN: So, you will use an algorithm with a certain degree of confidence. Those that fall out of the confidence will have to have a manual review?
MR. BROSEKER: On the part of those trying to enroll that provider, yes, some action going back indicating that we think it might be a duplicate and can you collect some additional information.
Now, if you say manual, this is going to happen on an interactive basis. That is the way the enumeration system is being set up.
It is not like they will submit a form and two months later they will get something back saying, we have a problem with the information you are supplying. They will know right on the spot that there is a potential problem.
DR. LUMPKIN: In the sense of simplification, are we talking about a system where the provider, provider system, the payer system, gets back, this doesn't match or this may be a match, provide additional information?
I am just trying to get a feel for, does this really make things easier.
MR. BROSEKER: There will be a range of messages that they will get back, depending upon how critical the error is in weeding out duplicates, making sure we don't issue more than one number.
Some matches won't stop a number from being issued, but we will ask that other information come back to us. Other matches will.
There will be a range, depending upon what the condition is, what edit criteria the application fails.
I believe -- Karen Trudel and Mary Emerson work for me. Does the documentation, the file formats, include some of the types of error messages that one can get back?
MS. TRUDEL: The error messages, yes, are included in there. The search algorithm is not public at this time because we are still working on it.
It is going to be similar to what the Social Security Administration uses to make sure they don't issue two Social Security numbers to the same person.
It has to do with the name and a range including the date of birth. We are looking at the Social Security number, if that is available.
We are also looking at a combination of the state and the state license number for professionals. We found in some cases that individuals with, for instance, compound last names can wind up enumerated twice, because they look at totally different people. But if you look at their state license numbers, it is obvious that it is a duplicate.
What we will do is rank all of these matching criteria and come up with essentially a match grade. If the match scores above a certain amount we assume it is a duplicate. If it is below a certain amount we assume it is a non-match.
If it is anything in the middle, then the individual who is going to resolve this will look at it. We will also give back all the records that are possible matches, so that the person who is resolving this will have all the information that we have available to them at that time. Does that answer your question?
DR. LUMPKIN: Yes.
DR. COHN: It is getting hard to know where to start. I am presuming, on a very basic level, that you are not going to be using numbers; is that correct?
MR. BROSEKER: That is correct.
DR. COHN: If he dies or stops being a practitioner, that number will be retired from service and all of that.
MR. BROSEKER: That is correct. We have no intention to re-use them.
DR. COHN: Just to clarify for me under types of providers, it says group members will be linked to group. I assume that is through the national provider file?
MR. BROSEKER: Yes.
DR. COHN: Or is that through the number somehow?
MR. BROSEKER: Through the national provider file.
DR. COHN: So, even though that is under types of providers, that is meant to be under the file piece.
MR. BROSEKER: Right. If I understand that correctly, you wouldn't be able to look at the individual's number and just by looking at that know that they belong to such and such a group. It will be through other data elements.
DR. LUMPKIN: If I could jump in there, that is going to make some assumption of the validity of that information over a period of time. People's relationships frequently change, sometimes even more than daily. It depends upon which deals fall through.
What is the requirement that is being thought of for refreshing that data, and is that a payer requirement? Is that a provider requirement?
MR. BROSEKER: It will be, for Medicare purposes, it will certainly be the Medicare contractors who do that. For other types of providers it will depend upon -- we use the term subscriber to identify programs beyond the Medicare programs.
Now will subscribers be any payer? Will it be some state entity? Will it be some type of clearinghouse that new providers will go to, to get a number?
That is not defined yet. We would expect subscribers to follow certain rules in making sure their data is kept up to date.
They will, of course, have to go -- if a subscriber knows, if they are getting information from a payer and the payer wants to make sure that that provider is no longer affiliated with their particular plan, they will have to send update information in relating to that.
Whether or not that has to be forwarded to the national provider system, again it depends upon whether or not that is a claims processing issue or an identification issue.
The national provider system -- okay, I have the draft right here saying, we require each health care provider to forward updates to the data in the data base to a health plan within 60 days of the date the change occurs.
Now health plan is their payer. Now, for Medicare purposes they would supply the information to the Medicare contractor, using the same process they used when they initially enrolled for Medicare purposes.
DR. COHN: This is obviously a committee discussion question, since we are sort of peppering you with questions.
Obviously there are a number of issues that I somehow want to waive so that this doesn't turn into an ongoing bunch of questions and we will see you again next year.
I guess I would ask the subcommittee -- maybe this is out of term -- how we should go about flagging issues that we need to somehow continue on with.
Are we going to be having Mr. Broseker back at a subsequent time after we have had a chance to review the regs? How should we proceed on this.
DR. STARFIELD: I think you should make good notes and we will develop a method of communicating what we need to know.
DR. COHN: In that case, I want to flag a couple of issues that I see. One is location code, which may be an area that this committee may want to investigate a little further.
I would adjudge, as I look at this, that -- I mean, I think you are right. It is certainly the most controversial.
Certainly, the current methodology you have for developing it really will require a clinician, at every separate facility, to re-register with Medicare effectively, since the location code, since it is only two digits, will not be a location specific code, but will have to be re-given to that clinician, one would imagine. Am I mistaken in that?
MR. BROSEKER: Not for those who already have UPINs. We will be able to identify that information initially. We have location information now for Medicare.
DR. COHN: Let's talk about Dr. Lumpkin, who goes to a new emergency room to work and he has a UPIN number, or I mean has a national provider identifier already and is in a new hospital now.
Now that hospital may have a normal identifier location. But since he had to travel 100 miles to be in that place, and since this is only a two-digit number, he may already have a local one that is the same location identifier, the same two digits. Isn't that possible?
MR. BROSEKER: The two-digit codes will relate to that individual's number only.
DR. COHN: That is right. So, in other words, each time the person moves they really have to get a new location identifier for each location they may work.
MR. BROSEKER: That is correct. It could be an additional location. It could be a replacement location. If they actually move and they stop using the first location, we will note that in the files and then activate a second location. They could keep both locations open.
DR. COHN: Yes, but my comment is that hospital X would not always have the same identifier location, for example. It would be unique to the individual.
MR. BROSEKER: That is correct.
DR. COHN: Effectively it means that if a provider went to a new location, that number and location identifier has to be reissued to the person.
It is just something to be aware of and I think it does bring up a whole host of administrative simplification issues, a quagmire.
DR. LUMPKIN: For instance, the mobile CT scanner that makes the rounds.
DR. COHN: Thank you for mentioning that as an option. Anyway, I would just bring this up as a wide open issue that we may want to go further into.
MS. COLTIN: With respect to information that may change over time and that you are planning to update, are you planning to maintain audit trails on that type of information, so that you would know dates on which things change?
Oftentimes, when we go back, looking at data over a period of time, you want to know at some point in the past what group was Dr. X affiliated with.
Will that type of information be available, or are you planning to overwrite that kind of information and not store changes?
MR. BROSEKER: No, we do intend to maintain historical data and store changes without overlaying previous information.
I won't say that applies to every data element that we have in the file. I do believe that the file formats that are on the internet identify which sets of data elements are repeated for historical purposes.
DR. MOR: One of the data elements, classes of data in the national provider file, is sanction information. I am assuming that is because somebody wants to be able to get information about providers who go from one place to another and may have been sanctioned in one location.
You said you would seed it with HHS-type sanctions but that presumably in the future other kinds of sanctions might be possible.
One of the reasons that nursing assistants in every state in the country are supposed to be enumerated and certified in some way through training mechanisms is precisely so that they can be cross checked against criminal records and those kinds of things.
Would you imagine that this would ultimately serve a national purpose, because right now that is only a state function?
MR. BROSEKER: I think that what we would look at there is what we are legally authorized to do. In addition to the Medicare program, there were also modifications made to federal procurement law that said, any entity or individual who is sanctioned by any federal program is barred from participating in that particular program and is automatically barred from participating in other programs.
That would be the extent to which we would apply that sanction right now. At least, that is the basis of our legal authority.
Whether we go beyond that or not would depend upon what the issue is and what our attorneys say about our authority to do that.
There are no plans to take that kind of information into account at this point.
There is a national practitioner data bank that is maintained by the Department of HHS as well. We will be talking and coordinating with those folks to make sure that we aren't duplicating things.
I am not really sure whether they collect any data involving violations of state criminal statutes or not.
MR. GELLMAN: I think you said that the national provider file would be made available on the internet; is that right?
MR. BROSEKER: We have information about the national provider file on the internet. The file format is on the internet.
We do hope to make information available on the internet as well; if not ourselves, maybe other entities will as well.
For example, the UPIN directory, which is published today by the Federal Government, is currently on the internet. It was put out there, I believe, by the University of Wisconsin.
So, that appears to be a mode that either we or others who get the information that we do release, are allowed to release, will get it on the internet eventually.
MR. GELLMAN: That is all the detailed information that you have in the file?
MR. BROSEKER: No, the UPIN directory today contains names, UPINs and zip codes of practitioners. We believe that we will be releasing more than that information under the NPI. We believe that address information will be released.
Beyond that, our release will be under controlled circumstances, beyond name, address and number. That kind of information, sanction information, for example, we do not plan on posting on the internet.
DR. STARFIELD: I am confused about the discussion about location codes. You describe a two-digit code. You also talked, in the context of that code, about a zip code, which would take care of mobile units, actually.
Why are you sticking with the two-digit code, and under what circumstances might you replace it with a zip code?
MR. BROSEKER: I think today physicians who practice in multiple locations for the Medicare program have probably multiple PINs related to those locations.
In assigning a unique identifier, we have to do away with those multiple PINs. They will no longer get PINs.
So, we just decided, for however many locations we identify via multiple PINs when we do the initial loads, we will convert that to a location.
Those multiple PINs have addresses and everything else you need to pay a claim related to those PINs. We will pick that information up and load it in the national provider file.
We won't call that other location by its old PIN. We will assign a two-digit code, which is an alpha numeric code, and does cover close to a thousand locations using alpha numerics.
Now since we intend to use this for Medicare processing December 1, 1997, we needed to something in order to identify locations and something very easy and simple that providers can supply to us in billing, along with the eight-digit number, so that we could find out what location they were practicing in when they rendered the service, so that they don't have to give us anything else. Our current formats allow for a two-digit code to come in.
Now, those formats are subject to change as a result of the claims processing standard. We will need to use at least the two-digit code, at least for Medicare purposes, until we know what the claims processing standard comes up.
DR. STARFIELD: Now we are into the confusion between billing and service again. You know, if you are interested in where the service is provided, it is different from where you are going to pay the bill.
DR. LUMPKIN: Certainly there are multiple locations in Chicago, for instance, where there may be a surgery center on the first floor and an outpatient department on the second floor. A nine-digit zip wouldn't help in that kind of location.
I do suspect we need to move on to the payer codes. But I would encourage you that there does seem to be a fair bit of interest on the committee of the fact that we seem to be linking the location code and the provider ID.
To the extent that we are thinking about moving toward the requirements of K2, you may want to think about how those may be linked as they are presented to the committee.
I could foresee some difficulty if we are asked to buy off on a linked provider ID and a location ID, even though there may be some very important transitional reasons why HCFA would need a location ID to be tied up internally.
MR. STREIMER: I guess one of the advantages of following Joe is the fact that the payer ID initiative has been following a fairly parallel track as the national identifier. My presentation tends to be a little brief because it covers a lot of the issues that Joe has covered.
On the other hand, the payer ID is not as far along as the national provider identifier. So, there are a number of issues that remain to be resolved. Again, we would welcome your input to helping us resolve those issues.
Let me begin with that, and let me begin again by thanking everybody for giving us this opportunity to present to you these activities.
The payer ID is our acronym or our affectionate name for the unique identifier for health plans and employers.
When we embarked on this particular initiative a number of years ago, we were only looking at health plans, or payers in the generic sense. We did not expand it to include employers until the enactment of Kennedy-Kassenbaum. So, we do have a lot of legwork to do in terms of rolling in, or folding in, the employer part of this.
What we are presenting, as with the NPI, is a proposal. We are not wedded on this particular proposal. We think it is a good idea. We have had a number of consultations with industry in terms of a unique identifier for health plans and employers. But we do want everyone's input, including the input that you could present today and down the road.
Payer ID, who does it cover. This was very explicitly covered in the new legislation in terms of who needs a unique identifier, who needs to be enumerated.
The list before you, while you could take a few moments to read through it -- I don't need to read through it for you -- are all the entities or organizations that have been listed in the legislation.
HMOs, I will point out that while the legislation, I believe, specifically says health maintenance organizations, we do look at that more broadly to include all managed care organizations.
To answer an earlier question, we will be enumerating managed care organizations with payer ID as well as with the national provider identifier. So, those organizations will get a unique identifier for both situations, depending on the need.
So, we will be covered in both cases. You are right. They are sometimes termed providers as well as payers.
What are the benefits to industry with regard to the payer ID? Well, first of all, it is a standardization of data.
We will now have a unique identification number for plans and employers that will be used throughout the health care industry.
It will replace all the narrative names and addresses that currently exist.
Just as an anecdote, for the Medicare program, for example, we have up to 400 variations on a single insurance company's name in our file.
For example, Empire Blue Cross and Blue Shield out of New York is referred to as Blue Cross and Blue Shield of New York, New York Blue Cross and Blue Shield, and every possible variation and abbreviation you can think of.
My understanding is that can approach up to 400 variations. So, you can see how complicated that can be from a systemic point of view.
It is also very complicated from a provider's point of view in terms of filling out a claim form, or passing on a transaction that requires this particular number.
We believe that by having this unique identifier, we can eliminate the need for a health care provider to include a health care payer, their name, their address, et cetera. All that information will be resident in a system, so that a provider of health care will only have to enter a number from here on in.
It also replaces other numbering schemes that currently exist. There is an NEIC co-code that exists out there. Insurance companies require their own payer enumeration schemes. We are hoping that payer ID will be able to eliminate all those existing schemes.
Another benefit is that it is a core data element for health care transactions. Again, as Joe pointed out earlier, what comes out of the other groups in terms of the claims transaction group, the remittance and payment group, all those other administrative simplification transactions that require standardizations, we are hopeful that the payer ID will serve to fill the need on all those transaction for a unique identifier for plans and employers.
One of the other things that the new legislation required is that we need to consider other uses for these unique identifiers.
While we may come here before you with certain biases, especially with regard to Medicare and Medicaid, we need to think more broadly in terms of how these identifiers can be used across the entire health care industry.
The payer ID in particular, we are looking at for routing claims between one payer and another, whether it is primary to Medicare or secondary to Medicare.
We think it is an important routing tool for coordination of benefits when other payers are involved.
We think it can also serve well as a mechanism for tracking other health plans and employers in terms of what is going on in their business, because all the transactions across the country that deal with that particular health plan and employer should have that particular health plan employer's number on that transaction.
The consultation partners, the list is pretty much the same as Joe had gone through for the national provider identifier.
Certainly the national committee is on our list. At the top of the list here, the work group for electronic data interchange, the national uniform billing committee, the national uniform claim committee, the American Dental Association, the American National Standards Institute, and certainly all of our partners and stakeholders and customers that will be involved with this particular identifier -- the states, the provider focus groups we have been meeting with an dealing with, provider associations.
We cannot forget about our electronic billing vendors and clearinghouses, those who serve the customers or those that serve the providers and payers to be sure that they are brought in early in the process, so that they understand what we are doing, and certainly all the federal agencies across the country.
Some of the specifics for payer ID, what is it. First of all, it is a number. We are looking at a nine-position unique identifier and I will talk a little bit more about that in the next slide.
It is also a registry. It is a place where someone can go to get a number. We have to build the system or the infrastructure that surrounds enumerating health care payers and employers.
I guess I would make the analogy to the national provider file or the national provider system that Joe spoke of in terms of the national provider identifier.
We will be developing what we call a registry, so that any payer or any employer who seeks an identifier has a place to come to, to get that identifier.
It is an electronic phone directory. It is an ability for anyone to know the address location for sending a claim, for making an inquiry, whatever the need may be, in terms of a payer ID. We will have a phone directory to serve that need.
It is also a system, an enumeration system. We have the overall capacity to enumerate 100 million plans and employers.
Right now we are estimating approximately 4 million employers based upon conversations with the Department of Labor.
I would like to say at most, but that is never correct, but we are looking at a potential 4 million payers or health plans in the country. That number will vary because we have heard different variations of it already.
A little bit about the structure of the number. At first it will be a totally numeric number. It will be eight digits plus a check digit for a total of nine positions.
What I have displayed for you on this particular chart is how we are considering issuing these particular numbers.
Based upon our consultations with the industry to date, we have learned that there are a number of large health plans out there who choose not to be enumerated with a nine-position number.
They feel that their customers are better served -- and we agree -- to have a smaller base number that basically represents their major corporation.
They can possibly use a suffix on that base number so that they may identify specific plans or products within their particular corporation. Let me give you an example. I may not have done a very good job in explaining that.
If you look at line one there, at that second line where we will be assigning numbers in blocks of 100, for a larger corporation such as Aetna, which has a number of health products and plans.
They may like their first six digits to be the same for all those products. So, anybody who looks at the first six digits can tell, this is Aetna. We know it is Aetna. We don't know what particular product of Aetna, but we do know it is Aetna.
We would leave it up to Aetna to take those last few digits and assign those to their particular product or plan.
There are many health care providers or health care plans out there that are small enough that they do not need a six-digit base number plus the three digits. So, we have also reserved some lines for those folks as well.
For example, line two is reserved for those health plans and employers who only need a single nine-position number. So, we wanted to meet the needs of both the large and the small plans.
As a result of meeting the need and retaining the nine-position identifier, we sort of are accommodating both by having a six-digit base for large employers if they choose to have one, or large plans, and nine-position numbers for smaller ones who only need one or maybe two numbers.
I also want to mention that there is one bit of intelligence in this number. Based upon our conversations with health care providers, they felt it would be very beneficial since up to 85 percent of the Medicare population especially have Medigap plans, that they be able to identify whether this particular plan or a beneficiary is a medigap plan.
So, we have reserved the last two positions, those last two Xs, if I may, 90 to 99 will be reserved for medigap plans.
So, when a beneficiary walks into a doctor's office and hands that doctor their card, their Medicare card, or their other health plan card, if there is a 99 at the end of that particular payer's ID, the health care provider knows right away that this beneficiary is enrolled in a medigap plan. That is at the response, again, of the health care community.
Some of the data elements we will be capturing for payers and for employers, we will have their names and some of their aliases, as we like to use, because again, some payers go by multiple names.
We will have their mailing address, their electronic address, their billing address, any other address they feel a transaction might need to be sent to.
I might add that, for those large plans who want to use a block of 100, they can assign a particular suffix to one of these addresses if they so choose. It will be their choice.
However, for purposes of administrative simplification, if a provider doesn't want to be worried about all of those suffixes, they may just choose to send the first six digits with a 00 suffix, and it would be up to the payer to figure out where that particular transaction would go.
I use the term transaction intentionally. We are not just talking about claims here. We are talking about all the potential transactions that can take place among payers.
If a provider, for example, wants to make an eligibility inquiry, that particular payer may have a different address for where they would like the eligibility inquiries to go.
It may sound complicated, what we are trying to do, but from the provider's perspective, if they just put a 00 at the end of those numbers, it is up to the payer to make sure that that transaction goes to the right place.
We will be capturing telephone and fax numbers about the particular payer. We will cross reference the number to other numbers associated with the payer.
We will also categorize the type of health plan that it is, based upon, generally speaking, the categories that are enumerated in the legislation.
We will have service codes. Many payers have different plans for their dental products, for their vision products.
We will be able to identify what payer ID is associated with which plan, because there will be service codes within the payer ID data base.
We will also have function codes, as I indicated earlier. There are different functions that are performed from a payer's perspective, whether they are checking eligibility on behalf of the beneficiary and the provider, whether they are checking enrollment.
There may be a different need for the payer ID and that particular suffix that I am speaking of, in terms of functions to satisfy the customer.
Some of our time frames, we believe that these are a bit ambitious but we are striving to achieve these time frames.
We hope to publish a notice of proposed rule making in The Federal Register last month. That is very ambitious. We have yet to move an MPRM out of the agency, but I am very optimistic and I kept it on the chart.
It is not that that is going to make it move any faster, but we will try.
One of the things that we are re-thinking, right now we have been looking at whether to use payer ID on a voluntary basis in the Medicare program, sort of as a test bed, where providers may choose or may choose not to use the payer ID on the claim form.
We have put out some instructions with regard to its use on the claim form.
We are now re-thinking this because we think it might not be the best idea to put in excess of 30,000 payer IDs out into the Medicare environment before a final rule is published, especially if, as a result of notice and comment, we change the payer ID and it is not adopted as the standard unique identifier.
So, we are re-visiting that. That April 1997 date is very much up in the air.
We are also looking to publish a final regulation in July of 1997, and requiring it or mandating it for Medicare in January of 1998.
Now, let me point out something just to clarify these dates as well as maybe a couple of dates that Joe talked about in terms of the NPI.
While the law says we adopt within 18 months of enactment, and that plans have up to 24 months after adoption to implement, those months will vary depending on when we adopt.
So, for example, if we were to adopt the NPI or the payer ID in February of 1998, which is like the last point in time we should be able to do it, then we would have until the year 2000 to basically implement.
Since we are looking at adopting these particular identifiers earlier than February of 1998 -- July of 1997 to be exact -- the implementation time frame is still 24 months, but it is really July of 1999.
I hope I made that clear, because the time frame begins at the time the standard is adopted.
The last bullet on this particular schedule would be that in July of 1999, the entire industry with the exception, I think, that small plans have a longer period of time, would have to come into compliance.
Some of the challenges that we are still encountering, we are learning more and more every day in terms of industry needs for the identifier.
Some needs are very different. We have tried to approach this with a very open mind, beyond Medicare and Medicaid, and we are learning a lot about how the industry feels about the identifier and what industry's needs are, and we are trying our best to meet those needs.
Timely adoption of the unique identifier is certainly a challenge before us. While the law give us until February of 1998, there is a lot of work that needs to be done before then.
So, we are still shooting for July of 1997 to the best of our ability. If that slips, hopefully we will have good reason for it slipping.
We want to ensure that the source data are accurate and includes employers and health plans. It is very important -- I think the point came up earlier -- in terms of how do we plan on keeping our data bases current.
With regard to the payer ID, it is going to be very much dependent upon the health plan and employer. Our plans are to give them direct access to our system, so that they can update the data when it needs to be updated.
We think that is very important, to be sure that we have it as current as possible.
Another implementation challenge is certainly to facilitate industry compliance. I think Jerry used the term or the phrase, one size fits all.
As he indicated, not everybody will be happy, because today there are a number of systems and numbering schemes out there and we need to bring industry along to see the benefits of having a unique identifier.
We also need to be sure that we make it as simple as possible, not only for the providers, but for the health plans and employers, so that they will embrace the system and they will get enumerated.
The status of some of the activities. We have had the payer ID approved from the USA Registration Committee. This is a committee that falls under the umbrella of the American National Standards Institutes.
This means that we have been approved to use payer ID on health care cards, should a health care card be used by a particular payer.
Of course, this approval came before enactment of the Kennedy-Kassenbaum legislation. Certainly if things change as a result of the notice and comment period and our consultation with industry, we would have to approach the USA Registration Committee again.
We certainly think this is a major step forward in standardization, to have their endorsement.
We have an interagency agreement with the National Technical Information Services. This is an agency of the Department of Commerce.
They are actually designing the registry. Initially they will be in charge of the registry to be sure that employers and plans are enumerated.
They have taken on a lot of the programming tasks, the public relations tasks, et cetera, and we have been working very closely with them.
We will continue our consultation with internal and external groups. Clearly we have a lot before us and a lot behind us in terms of our consultation.
As we learn more from the industry, I think it is only incumbent upon us to take it back to our consultation members and our consultation partners and bring some of those issues to the table.
We are working on implementation strategies. One of the strategies or one of the ideas that we are wrestling with is how to enumerate four million employers and four million payers.
I think where we ended up, at least initially, is that we are leave it up to them to come to us. We are not going to try to find all of them.
I think that is very important because I think the way the law reads is that each plan or employer who engages in electronic data commerce will be required to have a unique identifier on those transactions that they use.
So, we will be looking to the health plans and the employers to come to us to identify themselves as engaging in EDI, so that we may enumerate them.
The payer ID registry system is currently being developed. It is almost completed. We are reviewing work products of the NTIS right now.
Hopefully over the next couple of months we will have a lot of information and a lot of updates to provide the industry.
Right now we are in the process of collecting information from the NTIS on the payer ID data base, so that we can begin some of our testing.
I know that was a lot in a short period of time, but I did want to give you folks an opportunity to ask any questions that you might have.
DR. MOR: Could you just -- I may be dense here, but the interrelationship between employers and payers. Those would be separate sets of numbers, just to make the provision that employers could be the payer, or might have an administrative agent?
MR. STREIMER: The way we are making the difference, again because the law separates the two, if an employer happens to be a payer or a plan, they would only need one number.
We expect that there are employers out there who have other payers administer their health benefits for them.
Well, question arises to that employer, well, if my health plan has a number and they are uniquely identified, do I need a number.
That is a legitimate question and that is a question that they need to pose to themselves in terms of EDI that they deal in.
For example, enrollments and disenrollments. If they interact with their particular plan or plans on enrollments and disenrollments and it is electronic, they may choose to have an identifier that uniquely identifies them as an employer, so each of the plans they deal with has a unique identifier from which to deal.
If they don't have any EDI they may say, well, I don't need a number. My plans do it all. So be it. That is fine.
DR. COHN: I will start with one or two questions here. As I look at your payer ID, I am actually a little confused. I am just sort of wondering from both of you about the likelihood of a unique payer ID being the same as a unique provider ID.
It seems like there is a strong likelihood that a provider ID could be the same as a payer ID.
MR. STREIMER: First of all, they are two different sizes. The numbers are two different sizes. The payer ID is nine positions.
Your question is a good point. You notice when I went through the chart it showed all the Xs and you notice that the first line that I skipped over that begins with zero, will not be assigned.
The reason that will not be assigned is that when you look at this number, especially when it is put on a health care card, there could be that confusion, that when the first digit is zero, somebody may automatically look at that number and forget that it is a zero and assume that the remaining number is a provider ID and not a payer ID.
So, to minimize that type of confusion, we won't even be assigning that first row of numbers for payer ID. But other than that, it is the number of positions that really distinguishes the two.
DR. COHN: Can I ask one other question? This had to do with, I think, the differences between Mr. Broseker went to particular pains to discuss the issues of a meaningless identifier, where the meaning had been stripped out of it.
Obviously you are taking a completely different tack where you are providing a lot of meaning in your number, both about the payer as well as potentially the product. It almost begins to sound more like a product ID than it sounds like a payer ID.
Under those circumstances, what happens when the 100 numbers that have been provided to an insurer are used up. Does the insurer have an option to re-use those numbers, so that, depending on what year, the numbers might mean a different product? How are you attempting to deal with that.
MR. STREIMER: A couple of different answers to that question. First of all, I wouldn't say that it was an extreme in terms of us building intelligence in this number.
The only real intelligence will be the suffix for medigap, the 90 to 99 where we are reserving that particular suffix for medigap plans.
Very few users of the number will realize from the outside that a one means a large employer or a small employer.
I think the more important point is the fact that it will be randomly generated for the most part, except that we will try to satisfy a particular customer's needs when they would like the first six digits to be the same. It will still be a random number for that particular plan.
As you can see by this list, we have reserved rows four through nine, where we might have to assign blocks of 100 or we may have to assign single numbers.
The real customer in terms of, if I may, that kind of intelligence is the payer, not so much the provider.
If a payer runs out of numbers, a block of 100, we can assign another block of 100. We would like to not reuse particular numbers or suffices, although it is up to that particular payer how they use their suffices. I think I mentioned that earlier.
They will decide what 01 means, what 02 means, what 03 means. That will be up to them. We will encourage them not to use 02, use it up, decide, well, we don't need it any more, and use it for something else. There may be some confusion as a result of doing that.
We would ask them just to get another block of 100 if they needed more numbers. Those first six digits would not be the same on that second block of 100 numbers.
DR. LUMPKIN: Can I just follow the question? You are not really giving them 100. You are giving them 90. If they don't have a medigap plan, then they only get 90.
MR. STREIMER: That is correct. But if they do have a medigap plan, they would be able to use those.
DR. BRAITHWAITE: Of course, Simon beat me to the punch with my question. I did want to point out that this is a perfect situation that shows why we have a need for administrative simplification, where we have 10 times as many payers as we have providers of health care.
MR. GELLMAN: I am interested in your using NTIS to help design some of this. This is not a question but a cheap comment.
If NTIS is going to do for you what they did for IRS with the cyberfile program. My question is whether the use of NTIS has any implication for the way your developed system and the information in it will ultimately be made public.
Other agencies have used NTIS to distribute information to the public in a way that increases the costs and limits the availability of information and violates lots of interesting information policy principles. I am wondering if you have thought that far down the road.
MR. STREIMER: Yes. I want to comment on your first comment, if I may. My response to NTIS, when this issue arose, was simply this. I said, now you folks have a real opportunity to show this country what you can do, and that you can do it on time, a quality job and under budget.
The response was, we are going to do our best. But your point is certainly well taken.
One of the things that we have been working very closely with NTIS was in terms of how we make this product available to the product, that we do not see this as an opportunity for NTIS to make money.
We see this as an opportunity for administrative simplification. That is not just rhetoric. That is exactly the conversation I had with them.
One of the things I constantly reiterate with them is that this needs to be made available electronically through a bulletin board mechanism, if at all possible, so that anybody who has electronic access -- and hopefully the majority of folks will -- can retrieve this information for free, except for their cost of hooking up to this information.
I have also asked them to look very closely in terms of making it user friendly. Certainly when you are dealing with so many plans out there and so many variations or so many products within a particular health plan, I don't think there is any provider in this country who wants to sit down and look at hundreds upon hundreds of plans and addresses, to be associated with that particular plan.
I have asked them to take a real close look at looking at ways we can maybe do this geographically. Maybe we can simplify this in some manner so that a health care provider can look it up very quickly, put a number right on the claim form, and send it on without a lot of burden.
That is their charge, and I am waiting to hear, in terms of how they will meet that charge.
On the other hand, there is an entire industry out there that they see the opportunity for potentially making some money. There is no doubt about it, for selling copies of their directories, just like they sell manuals today for the Medicare program.
They operate through funding received or achieved through these mechanisms. They are a self-funded agency. So, I do see the potential for them to try to market and receive some revenue as a result of this.
In terms of providers of health care especially, we have asked them to try to find a way to do it for free. For subscribers, I will coin the term that Joe uses, employers and payers, those subscribers who want direct access to the system and to be able to update the system, there may be a subscriber fee for them, to help offset the cost to the Federal Government for administering this identifier.
MR. GELLMAN: Let me just make the point that there are lots of rules and policies and laws that apply to the distribution of information by federal agencies.
NTIS says they don't apply to them and that they can charge more money and make profits and do other things that really significantly increase the cost.
It seems to me that it is your obligation, in doing business with them, to make sure that the public isn't prejudiced by the fact that you have decided to use NTIS, and that the public has to pay higher prices or not be able to use data in ways that it could if it got it directly from you.
MR. STREIMER: I agree.
DR. COHN: This has to do with a question about the fact that most of the list of payers are also providers, at least as I define them, or at least as has been defined previously.
I am just curious about when a payer provider is going to use a payer ID and when they are going to use a provider ID.
MR. STREIMER: I guess we both can answer that question. The payer ID -- just as I may from an example point of view -- let's say on a claim form, there are specific requirements on a claim form in terms of the provider of health care. That is when the provider identifier will be used.
There are also specific elements on a claim form today that ask for whether there is any medigap insurance and what is the name of that medigap plan. That is an example when the payer ID would be used.
Your comment about most of the plans are also providers, I guess I am a little bit confused by. Besides managed care organizations, generally the list that I have shared with you, are all payers.
DR. COHN: I don't know. I guess as I look at Indian Health Services, Veterans Health Care, military HMOs, group health plans, a lot of them look like -- I was really questioning -- it becomes gray, is the comment.
I am just sort of wondering whether there will be some lack of uniqueness depending upon whether an entity decides to use a provider ID or a payer ID. It may cause some confusion. That was the question.
DR. STARFIELD: I think that is because we are still confused by what you mean by a provider.
MR. STREIMER: I think maybe for time's sake, I think we are going to make great efforts to try to be as explicit in the notice of proposed rule making, in terms of our definition.
I think we are obligated to be very explicit in terms of our definitions for providers and payers. Maybe we can try to address it then. I don't know where we are in terms of sharing some of that information at this point in time, in terms of the regulation. Have you had the opportunity to review the regulation at all?
MR. MOORE: I talked with Jim about getting the provider reg that is in the department right now going around for review, to the members of the committee.
MR. STREIMER: Good.
MR. MOORE: It is under the instructions that they are federal employees when they get that. I don't want to have them publish it before we get it through the process, because they are part of the review process, their comments.
MR. STREIMER: That should hopefully clarify some of that. I think for all intents and purposes, what we have listed for you today in terms of the payer ID presentation we envision to be defined as payers, not providers. Hopefully when you see that regulation, that will be clear.
DR. LUMPKIN: Would it be fair to say that if I am an individual physician who is working in an IPA model plan that is contracted with Medicare, that the physician would use the provider ID in billing to the plan and would direct it by via the payer ID for the plan.
Then the plan would then bill HCFA, whatever their arrangement is, using their provider ID.
MR. STREIMER: They would bill HCFA using their --
DR. LUMPKIN: We will simplify it; their intermediary or whatever, using their provider ID.
MR. STREIMER: I have to think about that one. I will say this. The provider will use his provider ID.
DR. LUMPKIN: Perhaps, since this will have to go through the full committee, if you could perhaps think about that and write us maybe a short little vignette that would explain the interrelationship, that would be very helpful.
DR. MOR: Regarding who gets a payer ID, I didn't see here, sort of the indemnity insurers, liability insurers, for auto accidents, fire insurance. Those would be under health insurance?
MR. STREIMER: They would get payer IDs.
DR. MOR: A new category of insurer.
MR. STREIMER: Yes. We will also have to look at situations where we have court settlements, et cetera, where we would have to enumerate, even maybe on a temporary basis for the short term need, those entities or individuals who would require a payer ID, so that they can receive payments.
DR. LUMPKIN: Would you include in that Workmen's Comp?
MR. STREIMER: Yes, I would.
DR. LUMPKIN: Any other questions? Certainly those of you who may have questions in the audience, I would encourage you to direct them directly to that.
If you think that you have comments that you think the committee ought to be aware of, we do have a time schedule, and I would encourage you to send those to him. He will share it with the rest of the committee.
This will have to go through the full committee, so there will be discussion. Certainly we would encourage you, if you have some input, either to send them directly to HCFA.
There will be a rule that will be published and open for public comment, or a proposed rule, and certainly the committee will be reviewing this issue as a full committee.
We are going to take a very short break, a 10-minute break, a chance to stand up, and then we will go back to our panel on health care purchasers.
(Brief recess.)
DR. LUMPKIN: Okay, we are going to go to the first panel. The first panel, because of travel arrangements and so on, we have the agenda with people listed on the agenda, and we are going to -- as I mentioned earlier, the nature of my clinical practice was in emergency medicine.
You always learned how to handle changing situations and respond to them as appropriate. So, because everyone is here for every panel, we are going to be a little bit flexible and I hope you all can bear with that.
So, for the first panel, again, this is a changing gears, the focus is on administrative simplification, the Kennedy-Kassenbaum legislation.
We have had initial discussion on the view from HCFA, particularly as relates to two identifiers. Now we want to begin to hear from panels of involved groups that look at different components of administrative simplification.
MS. BORDER: Here I am, Norma Border. I am national director for the National Association of Claims Assistance Professionals.
I represent a group that really serves two of the constituencies that you are most interested in today. It is a little daunting to find out that I am the sole representative for the American consumer this morning.
I thank you and the committee that you put the concerns of the American consumer first with regard to administrative simplification.
My constituency, my NACAP members, fully believe that administrative simplification can help make things better.
However, like you, they share current frustrations with the existing system.
The consumer essentially looks at the whole health care processing issue much like a black box. The RAND institute, back in 1989, said a health care consumer looks at a health insurance claim form as only slightly less complex than his IRS 1040.
When consumers are faced with the problem of submitting a claim or on their own with regard to filing a claim, they look at all those little squares and boxes on a HCFA 1500, or the Blue Cross form or the HIAA recommended form, and they make it through name and address.
They may get the Social Security number right, but then there is some ambiguous thing called policy holder or relationship to, they get a little confused about.
Then they get into provider. Even then they start getting even more frustrated. They are sitting there with bills from various, let's say, the South Bend Anesthesiologist Group.
Well, they saw Dr. Brown, the anesthesiologist, for 10 minutes the evening before the surgery. They didn't see him the morning of the surgery. They were too groggy already. They get a bill and they don't even know he was associated with South Bend Anesthesiology Group.
Then they are supposed to figure out, well, am I filing this on my hospitalization? Do I have to make sure this gets to my major med policy holder. They are as confused as everybody else.
Let me back up and tell you a little bit about NACAP's members. They represent two groups. I represent claims assistance professionals. These are independent consultants with extensive backgrounds in health insurance claims who are now working as consumer ombudsmen.
They work directly on behalf of the consumer to sort out the mess that falls out of the back of the insurers claims processing system, and they work to get a claim resolved.
My mom is a Medicare beneficiary who moved down to Florida. She opted for the HMO select with Humana, and she had to have her knee replaced.
Mom was just thrilled with the care she got. Two weeks after the surgery, my mom called me and said, well, you will never guess what I got in the mail today. I said, what, mom.
She said, I got this bill for $16,000 from the Humana Hospital for my knee. She said, I guess I had better mention it to the girls in the office when I go in next time.
I said, mom, when are you due to go in. She said, well, I see the physical therapist in two weeks. I said, mother, you have a typewriter in the bedroom. Sit down now and let me dictate to you what you write, because you are now a private pay patient. That bill has not been submitted to Medicare. They don't even know that your surgery was covered. They want you to pay that bill.
She laughed and she said, oh, no, they wouldn't. I said, yes, they would, mom. Trust me.
It turned out that the reason my mother was billed as a self pay patient, a Medicare beneficiary who had opted for the HMO program, was because she had gone to a primary care physician who had referred her to the orthopedic surgeon.
The orthopedic surgeon had scheduled her appropriately for the surgery, but he was going on vacation and he failed to file a follow-up referral notice to the HMO that, in essence, says, I have completed this surgery. Therefore, you should bill this to Humana HMO.
When that little referral notice did not go through the process, my mother became a private pay patient.
This is the kind of paperwork mechanism that is almost invisible to the average consumer. The average consumer does not understand gatekeeper. They don't understand the hurdles that they need to go through to make sure that, if I walk in the door with commercial insurance or I walk in the door with Blue Cross insurance, or I walk in the door as an HMO enrollee, that the system itself is consistently identifying me and making sure that my claims are effectively routed.
A claims assistant professional takes on the role of finding out what the procedures were, what wasn't done, and finds out from the insurer or the provider what needs to be done to correct the situation, so this individual, who is validly enrolled under health insurance, has that health insurance paid for, has the appropriate payer or secondary payer or tertiary payer or individual insurance policy make payment against this health plan.
It is going through this maze that so many consumers find frustrating, what to choose.
In some instances the consumer still has a choice with regard to health care. In a lot of instances, once you become employed by a particular entity, your choices are limited.
The mechanisms required by those choices, believe me, are not fully explained in the little employee benefit pamphlets.
The other thing, I always like to go into a group and ask how many people have read their health insurance policy or read their entire employee benefit program, and do they understand the appeals mechanisms when something goes wrong.
HCFA has spent years trying to educate the Medicare beneficiary that, yes, any denied Medicare claim can be appealed.
Yet, to give Medicare its credit, very few claims are denied. But of those claims that are denied, less than two to three percent are ever challenged.
One of the things I am consistently asked when interviewed is what can the average consumer do. The first thing is find out the appeals mechanism. Second of all, put your complaint in writing.
Most people assume that if there is a customer service representation number, all they need to do is call, explain to somebody what is wrong, and it will be fixed.
What they don't understand is that most of the little consumer representatives can look at the CRT screen and they can tell you that yes or no, the claim has been received and yes or no, it has been flagged for payment.
But they are not equipped to change any of the information involved with getting that claim paid, and they don't exactly know how to instruct the consumer when there is a problem or when there is a piece of missing information.
We look to administrative simplification, the use of consistent provider ID numbers, the use of consistent payer ID numbers, to help that along quite a bit. We are very enthusiastic about that.
On the other hand we are concerned that, as all these numbers get generated -- it used to be that if you picked up a hospital bill, there was a narrative that described what was wrong.
You picked up the bill and there was a narrative which said, this is the procedure that was performed.
For the consumer, the ability to look at a bill and say, I broke my arm, I had nothing to do with a case of pneumonia, they can't tell from that five-digit number.
If there is a series of procedure codes, they are not equipped to challenge whether or not that was, in fact, what was done.
Everything is being converted to a number. What we are concerned about is that, as we move forward on administrative simplification, that we also ensure that we maintain access for the consumer; consistent, reasonable access for the consumer.
I would say probably 96 percent of the American public would never need the services of a claims assistant professional.
But for that six percent, they have a claim go through the system and pop out the back door as uncovered or an unexplained denial or a payment that is significantly less than that consumer had reason to believe would be made for the services he received.
It becomes increasingly difficult for that consumer to say, why wasn't Dr. Jones paid in full. Why is it only 10 percent of what he submitted as the charge and why am I responsible for the balance. What is my recourse.
Why was this claim denied? I looked at my employee handbook and they told me there was physical therapy or speech therapy or emergency room services, and didn't tell me the caveat, well, if it was a medical emergency, we are going to look at it and see whether or not it was a bona fide emergency.
Nobody is providing that kind of information on a regular basis to the consumer.
Not only access for the consumer, but access for those entities that are trying to assist the consumer. The claims assistance professional is an independent consultant and is reimbursed for their services.
The dollar that the health care consumer pays for a claims assistance professional, he is probably getting back $10 to $100 for each one he invests in getting a CAP to sort out the mess for him.
Our members deal with catastrophic situations. They deal with the double income family dealing with dual insurance. They deal with families where there are chronically ill patients.
They work on behalf of AIDS patients. Yes, they work for and with Medicare beneficiaries, to make sure that those individuals are getting full benefit from the existing health care coverage that they have.
I also have a significant portion of my membership that are working on behalf of the small to medium-sized physician practice.
This is the little doc. These are the partners. These are the five or less physician groups that are struggling to stay in compliance with filing requirements, that suffer rapid staff turnover, and electronics claims professionals are in there taking over the burden of their health claims processing.
They come in, pick up the bill, they convert it to the electronic format and see that it is directed to the appropriate payer based on the information that was captured on the patient ID form.
In both instances, in reviewing the legislation, we are a little concerned that neither one of the groups I represent, and their service to consumers and doctors, aren't clearly enumerated at this point in time.
Perhaps my electronic claims professionals are going to fall into a provider category. But my claims assistance professionals, working on behalf of the consumer, need to be able to go back to the provider and the various payers -- and if they could do it electronically and say, there is a problem on this claim.
I have spoken with staff at HCFA. There has even been some consideration given over the last year or two to the idea that if a Medicare beneficiary wants to initiate an appeal, it could be done electronically because you can identify the claim control number, you have the provider ID, you know who the intermediary was.
It would be possible to go back in the system and say, you know, I disagree with the result on this. I would like to challenge it. So, it would eliminate a lot of the paperwork.
That is contingent on our being identified in this loop. Both my claims assistance professionals, on behalf of the consumers say, yes, if you need to identify me, my hand is up, I will fill in the paperwork.
I would love it if I could start working direct to Medicare carriers and intermediaries to help my clients solve their claims filing problems.
That is only the tip of the iceberg. If I can now start going direct, or using the information administrative simplification is supplying me, like payer ID or provider ID, that allows me to interact with Blue Cross or interact with Aetna or interact with a managed care organization to find out where there is a problem on a claim, then we are expediting things for the consumer and making sure that they understand what they are entitled to in the way of access and appeals.
Right now the whole appeals issue, what do you do when a claim doesn't get paid, gets very confusing.
In working with managed care organizations who are both provider and payer, we have already encountered instances where the individual went to the "network" facility.
Because the administrative procedures within these newly designated organizations haven't been refined yet, they literally cannot tell the health care consumer, I don't know why that wasn't paid. The payer portion of our entity isn't recognizing what the provider portion of this entity is telling them.
So, the consumer sits there and goes, so what do I do. I am supposed to not have a deductible to worry about. I am supposed to only have a co-pay. Now you are telling me I am a private pay patient because you guys can't match the paperwork together internally. It just burgeons beyond that point.
We did respond to the four questions and I will go back to that. But I had to make our bid for the consumer right up front.
NACAP has been working with WETI and we are aware of recommendations that have considered through the various ANSI committees.
We strongly support administrative simplification. We know that there are a lot of vested interests. We know that everybody has their own number. They have been looking at it for years. They are very comfortable with the way it looks.
If you tell them, instead of being one, two, three, I now have to be nine, eight, seven, they get upset about it.
I think the committee could serve a big function to kind of like cut through some of the gordion knots and say, we are sorry, but if the world is to benefit from this, the whole health care community is going to benefit from this, everybody has got to be willing to back off from their own little numbers and open up and use one that means the same thing to everybody at the same time.
If I am facing a situation where an entity is both a provider and a payor, that I have some assurance that the number I am using will distinctly identify which part of that entity I am trying to talk to, consistently, all the time.
One of the major stumbling blocks if commercial insurers that have multiple field offices around the country that serve particular employer contracts.
If you cannot direct that employee claim to the correct billing office, it kind of disappears into limbo. The idea of a payer ID that would tell me I am always addressing this particular facility at this location and they handle these claims would be a boon beyond measure.
We are concerned as to how much of the systems requirement and administrative simplification requirements could adversely impact a group that are essentially small business owners.
They want to be involved in the numbering system. They want to have access to the directories. They hope that access to the directories isn't going to be thousands of dollars a year in subscription fees.
Like everybody else, they would like to be able to get it from the government for free. They also accept reality that there has got to be a way to produce it and provide it in a media and support the cost of the media.
They are willing to do this, if it keeps them in the loop and involved in the process and able to access and resolve issues on behalf of their clients, whether they are consumers or physicians.
At this point in time, essentially, claims assistance professionals, consumer ombudsmen, and electronic claims professionals who are in essence providing a billing service, are really invisible in the system.
The electronic claims professional is using a provider ID, under which they submit billing to the various payers.
The fact that it is going through a billing center that is capturing the data for anywhere from five to 100 small doctor practices is essentially unknown.
The only entity that might be able to capture that data right now is clearinghouses.
The claims assistance professional is initiating all their activities on behalf of, and providing correspondence and input under the direct authorization of their clients. They are in essence invisible to the system, too. But they are there.
We have claims assistance professionals who have been in business for over 20 years. We have electronic claims professionals who have also been in business for 20 years.
But every time I talk to everybody, oh, that is a neat idea and how long have they been around. Is this new. It is not. They have been there attempting to resolve the problems for the consumer and the small practice probably since shortly after the inception of Medicare.
They are a growing group. They have been able to reach the small doctor practice at a point in time when neither the major insurers nor health care financing has been able to make electronic billing attractive.
They have stepped in and been able to market to this particular niche.
In regard to question two in regard to the process being undertaken to determine this, we realize that there are some really stringent time frames that the committee is dealing with and DHHS is dealing with.
We believe that sticking to stringent time frames for establishment and adoption of standards, in essence, is going to pull some elements of the health care industry along by their noses.
In essence, it is going to be the one thing that is going to force a group that has been working at consensus but not always achieving it for like the last 20 years to sit there and play put up or shut up.
On the other hand, we are concerned that once we get past the adoption standard point, there is some recognition in the legislation for the problems of dealing with the smaller entity.
We would not want to see an implementation schedule so tight that it in any way creates fall out.
We are aware of situations now where there are a lot of small physician practices that are using one or another of a particular kind of software.
They are dealing with small vendors. They are dealing with other small business entities. They are not dealing with Microsoft.
They are not dealing with Intuit. There has got to be a time frame there that allows the vendors to accommodate the changes needed for the clients that they serve.
We hope that, wherever possible, some of the implementation requirements can be a little bit mitigated for the little guy, because he is using this month's billing revenues to cover systems upgrades, and the doctor doesn't want to be zapped with an additional consulting fee or charges to him for his billing center to upgrade his software.
He wants that at kind of an even rate, or he is just going to fall out of the system and go back to paperwork. There is a finite dollar figure and a time frame that could make the difference on that.
I think I already answered number three, what are the current problems. It is the inability to accurately direct claims to where they should go, for both professional claims assistance professionals and electronic claims professionals.
Do you realize that if a claim is submitted electronically and your software says that it has been sent, that there is still no guarantee that the payer does, in fact, have it?
There have been all kinds of glitches at both clearinghouses and major payers where claims, routed correctly, kind of fall into a black electronic box, because that particular entity is having software implementation problems and there is no mechanism for feedback.
There was a major player in the midwest that went closer to 45 days, seemingly accepting electronic claims from around the state.
At the end of the 45 days, when none of the physicians in the state were getting any payments back from that particular entity, the upper management had to become involved and go around practice by practice, to apologize for the fact that they hadn't informed anyone that there was a problem, that they hadn't issued any error codes and that they had, in fact, materially impacted those doctors' accounts receivable, payables and cash flow for almost a 40-day period, with no apology.
You know, this is the type of situation where everybody goes, that can't happen, the claims are going to go in and they are going to get paid. That is still going on. Administrative simplification would greatly assist that.
Information is improving. Again, we hope administrative simplification is going to even expand on that. Our members do not create or generate clinical data. They are merely taking the information that is being supplied by the provider or by the consumer.
Quality is improving with increased electronic claims filing. It could be better as a result of administrative simplification. That is about it.
DR. LUMPKIN: Thank you. Questions from the committee?
You mentioned about the provider ID. You discussed the issue of having a facility ID. Was that separate from a provider ID?
MS. BORDER: I use provider and facility almost interchangeably. It could be a physician or it could be a nursing home or it could be a home health agency. It could be a practitioner who is physically located at a facility.
I realize that he has the capability to bill separately from the facility. So, provider and facility are interchangeable, depending upon what the service requirements are.
There are physical therapists who bill through a particular hospital facility. But there are patients who they see at that hospital facility who represent their own client.
I can see where something is going to come through as an outpatient bill from St. Mary's Hospital, and there is something else that could come through from Joan Smith, physical therapist.
DR. LUMPKIN: You were here earlier, I think, when we had the discussion about the location code and the provider ID.
The issue that exists there is, I am Joe Smith. I am at my office and I work at a free-standing surgery center and I work at a hospital. I do a simple outpatient procedure that I could do at any of those three locations.
Would you believe that that ought to be one provider ID? Now there may be a facility fee charged which would be billed separately, but would there be one provider ID or should that provider ID be modified for the location at which the procedure is being done.
MS. BORDER: If I provided the service, regardless of the three locations, I should have a distinct provider ID number. Then I should have a location number.
DR. LUMPKIN: Not linked to your ID number.
MS. BORDER: Not necessarily, no.
DR. IEZZONI: Thank you for your comments. It was refreshing to hear the consumer's view. One of the things that you noted was that consumers are concerned that, when they get a bill, all they get are a list of numbers and they don't know what those numbers mean.
Obviously, ICD-IX, for example, has software that translates numbers into -- well, I will put the words English in quotation marks, English translations of what those numbers are supposed to mean.
What this would suggest to me is that where you might want to go with this is that if we do have a provider number, a location number and a payer number, that you would want to have some software that would translate that into English words or other languages, that the consumer might be able to understand. That does present a bit of a problem.
MS. BORDER: But it is not one that the system has to handle with administrative simplification. It is the providers.
DR. IEZZONI: Let me make a point why it could be a problem. I work for Beth Israel Deaconess Medical Center, but we are owned by Care Group.
The consumer that comes to our hospital might not know us at Beth Israel Deaconess Medical Center. I actually work on the east campus of that medical center. They might know us as Beth Israel, because we just merged with the Deaconess across the street.
So, when you want the consumer to understand who the provider is, you are going to have to make sure that the software has English in it that consumers relate to.
That is where I foresee the problem, that the linkage of the English translation to the number might not choose the English words that consumers would understand.
So, it might be very confusing for consumers, when they get a bill from Care Group when they thought they got their care at Beth Israel Deaconess Medical Center. It is just a heads up.
DR. LUMPKIN: As I understand it, is that an issue with administrative simplification, except to the extent that it might be more feasible to have software to do that, if the number of code systems that they are trying to convert into English are limited.
DR. IEZZONI: all I am saying is that we need to keep the consumer in the picture. As we are thinking about administrative simplification and getting excited about unique numbers, we have to realize that consumers, who are at the end of that service chain, or food chain, might not want to know the number, but who that person is.
So, there needs to be some way to get to the consumer through some software somewhere.
MS. BORDER: I can be converted to a date, a Social Security number. The care that I receive and the identifier and the procedure and everything else can be nothing but a long string of numbers, which is essentially an ANSI transaction.
If something is wrong with any digit in that entire string, which is probably as wide as this room if you look at it, and I as a consumer have to go back and figure out why I am looking at a $1,000 bill, I need something that will help me retranslate the string of numbers into something I can understand and tells me whether or not I go to my provider to straighten the mess out, or I go back to my Blue Cross/Blue Shield plan, or I go to my managed care organization and say, what went wrong here.
So, we are all for speeding up and converting the numbers in there. But no matter what we do, sooner or later some poor consumer is going to be standing there saying, why didn't this get paid.
They have to have enough clues in the system to help them go back to the appropriate entity and find out what went wrong.
DR. STARFIELD: I gather that is at least partly what you meant when you said that they have to have access to directories that are reasonable cost, if any.
DR. BRAITHWAITE: What proportion of the claims that your consumers have problems with have to do with the coordination of benefits between different payers?
MS. BORDER: Lots.
DR. BRAITHWAITE: Is that the majority of the problems?
MS. BORDER: No. Probably the first, let's say 60 percent of problems would probably be resolved as clerical errors, missing information, stuff that the providers didn't send along to the payers. There wasn't enough information just clerically for the payer to say, yes, I can pay this or no, I can't.
Probably there is another 30 percent that falls into the category of medical review. Is this a service that I pay for or is this something that needs to be reviewed or needs more clinical information in order to review it.
Is there a service cap. Is this a bilateral procedure that wasn't coded right, et cetera, et cetera.
Then coordination of benefits, when it works right, works beautifully. When it is wrong, it is let Mikey do it for an extended period of time.
Again, the consumer is the one who, without the information, is supposed to prove that somebody is wrong. The birthday rule is great, but too many people don't understand the birthday rule; in other words, it is not the old eldest parent, the parent whose birthday occurs earliest in the year.
I can consistently find you employee benefits departments and even insurance companies that do not understand the application of the birthday rule.
Therefore, they don't know that John Smith's son's bills should be sent to his wife's insurance company.
There are situations where the managed care organizations and HMO's for a long time always assumed that they are primary payer.
They have now discovered that they don't have to be primary payer, if they apply coordination of benefits correctly. They can, in fact, say oh, yes, you are an HMO enrollee, but by rights, this should have been billed to.
Coordination of benefits becomes confusing when the consumer attempts to exercise some choice in provider. They have a pediatrician that they are happy with but now they are into a managed care group and they figure they can go ahead and bill it on the husband's insurance, unbeknownst to the birthday rule which says that they should have billed it under the wife's.
Then the managed care organization says, no, you can't do it because that is going outside the network. They are saying, but I have insurance under my husband that would pay for this service and you are telling me I can't go there because I had to go there.
There is also the consumer that sits there and goes, hey, I got two insurances. I am going to bill both of them.
They do not understand that coordination of benefits reduces health care costs for everybody when it is correctly applied. They figure they are entitled to it.
If she has got Aetna and I have got Blue Cross, by god we are covered on this and cash to spare. This is an education of consumer. It has nothing to do with administrative simplification. But it complicates the issue. It is their perception of what they are entitled to as a benefit.
DR. BRAITHWAITE: I was actually reviewing your written documentation and was looking at your answer to question 3-A, about the current transactions and problems experienced.
I just wondered, just to I guess pull from you a little more of your thoughts on this issue of the NSF and the ANSI standards and whether, from your view, and especially those of you in your group who are the ECP members, who are actually the ones doing the stuff, what the views are in terms of the variability or suitability of those two, and how it impacts.
From here it appears that the main issues are lack of standardized code sets. Other than that, there really isn't an issue.
MS. BORDER: Essentially, if they have been billing for an extensive period of time, the odds are that they went to the national standard format.
Now they have vendors who are making the ANSI formats available to them. It is like, yes, they want to move to that.
But they may be in a situation where the Medicaid provider in the state is not prepared to support ANSI yet. Even under NSF, there are various payers in the state that say, yes, I will accept an NSF format. They are still issuing pages and pages of bulletins saying, these are the codes that you have to use in these fields.
It means that either the physician's office or the billing center are maintaining binders of okay, yes, we can go direct to Blue Cross/Blue Shield for X. But if we are submitting these claims, we have got to remember that we use these codes.
Sometimes it won't be just designation codes or payer codes, but it gets down to which codes that particular payer wants to see for certain services.
So, it means that, although the provider is being told by the payer, you must submit this code for this procedure, although there may be a range of 5 and HCFA is accepting 4, we will only accept one, and we will reject any that come in any other way.
Or, if you send it in, our software will bundle or unbundle it to meet our own needs. So, the physician may be in a position of saying, I know what I did and I feel I am selecting the most appropriate CPT code. But the payer is telling me, I am sorry, you can't use that one.
That is another element in administrative simplification that says, all right, payer, you may have criteria for paying for this service, but is it appropriate to dictate back to the point of billing as to what is a valid code, if the code is valid and included in the entire CPT.
They are sitting there helping their doctor make sure that what he is submitting is acceptable to that insurer, that payer. It is kind of arbitrary. Yes, we have a standard format, but no, we don't have standard billing.
DR. COHN: Can I follow that up with one more question? I certainly hear what you are saying about that. John at the beginning talked about issues of variability and how much variability is desirable and necessary, versus how much is too much.
From those comments, what is your view and the view of your organization about the required necessity of variability versus standardization in all this.
MS. BORDER: I feel that if you are going to be working with standard data elements, that everyone be required to recognize the validity of every standard data element.
That means that a payer may have some coverage distinctions they prefer to make about certain CPT codes. But they cannot go back to the physicians in the area and say, don't bill it. I won't accept it if you bill it. They can't have edits and plates that won't allow the submission of a bill with that code on it, or with that ICD-IX-CM code on it.
In other words, if you take CPT, you take all of CPT. If you take ICD-IX-CM, you take all of it. If you take the use of HCVCS(?), you take all of it.
Then within the payer you make whatever judgements are required for the policy that you intend to pay. But you don't sit there up front and say, okay, if you are billing this one you can only do this one and that one.
So, in other words, everything is valid as a data stream, and then what happens to it after the fact or at the payer is at their discretion. They are the ones that are writing the check.
When they sit there and say you can't submit something, then they say, I have created my own standard for what you can put on the bill. Then it is not standard any more.
MS. MOORE: I was going to ask, the codes that the payer requires, is that really something other than the service that was really provided by the provider?
MS. BORDER: It is a situation where you may have therapy series or you may have mental health series or you may have certain surgical procedures.
I mean, HCFA for a long time had this whole open biopsy issue. There was no way you could submit an open biopsy to HCFA. It just wouldn't fly.
People, therefore, would not submit open biopsy codes to HCFA because it was a waste of time. The thing is, this is a valid code and it should be submitted. The payer has the right to designate what they are going to pay for. But all the codes should be valid.
The burden on the small provider, on the small facilities, when they have to maintain six separate binders in the office and attempt to keep their staff trained on the contents of those binders, if she has a HCVCS and a CPT and an ICD-IX-CM and they are all valid, then the decisions on what is happening with the claim occur at the payer where they should. It doesn't mean that everybody is sitting here wondering what the edits are.
MR. MOORE: I have one more question with the coding. What coding systems do you find predominantly used out there to identify, because that is another standard down in this group.
What are the current systems used by the payers, since you deal across the wide range of payers, including HCFA, the Blues, the managed care organizations.
MS. BORDER: It is pretty much CPT, ICD-IX-CM and HCVCS. There are payers that, for some reason or the other, want something different, but it is usually like their own local codes. It is not like they are sitting there adopting a whole other coding system.
I know there are some pockets where that is a little bit different. But nationally, that is pretty much what they see.
MR. SCANLON: I wonder if I can ask, just from the experience of your organization, to what extent does the whole body of claims processing now more or less make use of electronic data interchange, again differentiating between the hospital claims, plan claims, and say small claims.
MS. BORDER: It is improving, thank god, and it has improved markedly in the last three to four years.
Of course, Medicare is way ahead of the pack with both 98 or 99 percent of institutional claims and probably 70-something percent of physician claims coming in electronically.
It is our understanding of that, that it is the larger group practices, clinics and facilities that are submitting the bulk of claims.
There is still a problem with small to medium-sized physician practice.
Commercial insurers have gone from probably 12 percent to just under 20 percent of electronic claims coming in.
Actually, I got some -- ASAM(?) publications just did a survey of their medical office billers and had some really good data about the level of commercial Blues and Medicaid claims that can be filed directly and filed directly has gone up a good 10 to 15 percent from what it was three to four years ago.
So, there is a growing acceptance. Then you have a state like Kentucky that says, even the Workers Comp is going to be electronic and kind of shake everybody up.
So, they are jumping on the bandwagon and it is improving. There are still a lot of situations where you can hear someone say, well, I can bill all your claims electronically, and this is essentially what people are hearing from vendors of software.
Yes, they put them into the computer, but there is still a significant portion of commercial insurance claims and special service claims that even for the Blues are being dropped to paper, put a stamp on and mail.
There is still a pretty good road to go before it is where we would like it to be.
DR. LUMPKIN: There has been some reference to a document that you handed out. We also have it on disk and we will have it on the web page for the work group, which is listed on the back of the agenda.
So, for those of you who don't have a copy and would like to read it, that will be available there.
I do have one final question which you should not feel obligated to answer today, but I would be interested in the answer.
You mentioned that your group represents individuals who work with small practitioners' offices, some of which may be affiliated with larger managed care entities.
I think it would be helpful if we could understand some of the cost implications of non-simplification, some of the short-term cost implications of moving to simplification, and the long-term cost implications.
Obviously, whatever system we move to will be somewhat different to each of those ones, and there will be a conversion cost. If it is anything like my wife's office that runs on DOMS, which is no longer supported by any living entity and in the spiritual contact with those programmers, obviously there are some significant incremental costs, but others may have some incremental costs. If we could have some rough estimate of what your group believes those costs would be, it would be helpful.
MS. BORDER: I will come back to the committee with that. Thank you for letting me represent the consumer this morning.
DR. LUMPKIN: Thank you. At this time we are going to move to our second speaker, who just walked in the door.
MS. KOSS: Good afternoon.
DR. LUMPKIN: You are not here as a provider of information services, but as a purchaser of health care.
MS. KOSS: Exactly. Good morning. I thought i was going to be speaking this afternoon. I want to say thank you for inviting IBM to speak as a purchaser, on the administrative simplification requirements of the health insurance affordability and accountability act. It has been a while since I have been in this forum.
I thought I would briefly talk about IBM as a purchaser, sort of give you a sense of what our health benefits program is like, and some of our primary goals, et cetera, and then speak briefly to your questions and then give you an opportunity to ask me questions.
I am Shannah Koss from IBM, in the event that anybody didn't know that.
First of all, IBM's health benefits program has really involved from an entitlement program prior to 1984 to a competitive and integrated health care management model.
Our framework for health care purchasing recognizes that health benefits must be analyzed and managed to meet four main objectives: cost control, quality, choice and strategic flexibility.
To pursue cost control, we try to optimize risk management through a number of plans that are both self funding risk, share, capitation and fully insured agents.
All of these plans really need to be affordable in order for our business to be competitive in our marketplace.
In terms of pursuing quality, we utilize health care management and measurement systems, specifically NCQA and HETUS.
In terms of choice, again, for us to be competitive in attracting the employees we want to attract, we offer a broad range of plans.
Finally, our overall benefits program needs to have strategic flexibility, so that we can monitor the marketplace and the legislative environment, to enable adaption to those changes that occur regularly.
One of our key mechanisms or approaches that we use to advance these objectives is our participation in the national HMO purchasing coalition, which is 10 of the major U.S. employers, purchasing together to leverage better prices, drive care evaluation and influence care quality, by using that market leverage actually to get improved information.
We conduct ongoing plan evaluations, that relies on the existing information streams that are often disparate or non-existent, but we use what we can get.
Standardized transactions like those envisioned in the act will help improve underlying information that supports reporting on costs and services.
With that background, I will turn to the questions that were outlined in the invitation, the first regarding expectations for the impact of the act, administrative simplification requirements.
IBM is a very strong proponent of automated standards that are consistent with existing industry standards, and facilitate comparable information across plans.
Inconsistent handling of transactions identified in the act are a source of escalated health care costs and a potential source of patient dissatisfaction.
Standards that support plan comparisons regarding claim information and quality will support our objectives as a purchaser and enable plan participants to make more informed choices.
Negative impacts, however, will result if the standards inappropriately limit information or mechanisms for facilitating benefits administration.
Your second question asked about our concerns regarding the HHS process. IBM doesn't have any immediate concerns with the process the department has put in place to carry out the administrative simplification requirements of the act.
Our concerns really focus more on the outcome of the process, and the potential for standards to significantly diverge from current industry standards, or create barriers to innovation.
One example of this is the emerging trend in the health care industry that our company will be piloting in terms of the internet for on-line enrollment.
We would want to be sure that any adopted enrollment standard would not preclude such an approach or limit how it is done, particularly as it is really being experimented with and at a developmental stage.
Your third question went to major problems with current transactions. Again, IBM does not have any major problems with the nine specified transactions in the act, in part because our third party administrators and our HMOs really carry out most of those transactions with the exception of enrollment and first report of industry.
As mentioned before, we do have several concerns that the absence of standardization in these transactions increases health care costs and can be the cause of patient satisfaction.
Wearing both my IAT hat and purchaser hat for a moment, standards should facilitate consistency of transactions without negatively affecting other information or technology requirements.
The standards ideally should focus on content and organization of the message, not the transportation technology or presentation of the data.
An additional area of concern that may fall under the provision of other transactions referred to in the act, are inconsistencies and inadequacies in quality or plan evaluation information.
In discussion with HCFA regarding federal actions that would support purchaser information needs, we indicated that a national focus on consistent and improved quality data would support efforts like the purchasing coalition.
Then lastly, achieving administrative simplification goal while accommodating the stakeholders, I guess from an IBM perspective, we recognize that not all transactions fall within the specified list, as I mentioned, the quality. In fact, some of those may not be ripe for standardization.
Of course, we recognize choosing one standard will always create some amount of concern, and ideally you want to select standards that cause the least disruption and require the least investment, while achieving your goals of standardization.
The main constraint that I would encourage you to heed or consider as you select standards is the one that I noted in our joint hat, really, which is the need to avoid creating barriers to new technologies, particularly those that improve user access or promote competition across the technologies that deliver the information and, in fact, lower transaction costs.
DR. LUMPKIN: thank you, and thank you for recovering so quickly upon walking into the room. To start off the questions, I wonder if you could help me understand your fears about barriers to innovation.
Particularly the issue with regards to the internet, I guess I don't understand how what has been discussed and particularly being discussed by some of the standard development organizations, how the format and content would affect the mode of transmission and utilization of those devices.
MS. KOSS: The hope is and the belief is that it does not have to. The concern would be that in creating certain types of transaction standards, that they reach to some of the -- they conflict with some of the modes of transmission that would be at the interface for the user.
I don't pretend to be the most technically astute, and I know you have some other IBM folks who will be talking to you that are much more technical gurus, shall we say.
The notion that once you adopt a standard, that it is completely exclusive, or must be exclusively used, raises the point, at what point does the standard end. Does it touch upon the technologies that deliver that standardization.
The hope is that it would not be. But in looking at the very detail -- and I have had a few conversations, of course, with our technical teams -- there is not a bright line, necessarily, in terms of when you adopt a particular standard and how it would affect the different modes of transportation of that standard.
It really is just encouraging you to be aware of that potential and highlight that it doesn't reach to that end, and that the user interface can be flexible, or the technology supporting the transaction would be flexible.
MR. MOORE: I have a possible situation, and Bill is aware of this one. HCFA met with a number of the VANS last week, and you know we are doing the NTS, and the enrollment and eligibility and those kinds of transactions, not looking at the claim as being one that could possibly be sent over the internet.
If you are looking at an internet connection, you are looking at HTML, and in doing that, do you really need to have X12 or another standard put in between that connection, when you are working in that environment.
If you are queuing up the data at one site in order to transmit it to another, then perhaps another standard that we would come out with, like X12 or something else of that sort, would be appropriate.
If you are gathering the data in an interactive mode and you are bringing it in, then you don't need that particular standard for moving that data from point A to point B.
I think as technology moves forward five or ten years, will we move so rapidly that we obsolete some standards before we get them established.
MS. KOSS: One of the examples that I was given internally and I know the concerns are broader than this, but some of the concerns raised in the debate between the HL-7 versus some of the X12 standards raises the same kind of concern.
There is a whole clinical set of standards and interactions that will interface with the transactions that are identified and that are not identical.
You don't just want to create something that is going to interfere with the ability of existing interface engines and the like to translate between those, you know, to again draw a bright line.
I don't want to pretend that I can give you the ins and outs of those technical concerns. I know for the open enrollment question, we not only have an offering, but we ourselves are looking at doing some of that internet enrollment.
At what point does your enrollment transaction begin, and would it preclude a browser based HTML enrollment transaction.
DR. LUMPKIN: Let me perhaps suggest that rather than putting you on the spot and that is it, we are engaged in a process.
I think it is certainly my belief and I am sure others of the committee that, you know, we could just imagine a committee somewhere inside Microsoft saying, well, you know, we only have to worry about 640K because that is all anybody will ever use in their wildest dreams.
But somebody sitting outside is saying -- we would like not to be in that situation as far as we can see. So, any suggestions that you have, if you would like to submit them to us in writing, I think that would be useful to the committee.
DR. COHN: Ms. Koss, I actually haven't met you yet, but I really appreciate your comments. I am sort of reminded, as you are talking, that I think most of us recognize that most of the standards that we talk about are typically a combination of what one might describe as an envelope as well as the data content.
This is something that the subcommittee, I think, has talked about from time to time. I think you are perhaps directing our focus to the important piece, which is probably the data content piece, and observing that maybe the envelope may be a little less important as we move into the future.
I think that is what I am hearing from you. So, I want to thank you for that reminder.
MS. KOSS: The definition of it shouldn't be so hard and fast that it drives the technologies or the interfaces that link the two.
DR. MOR: I would like to -- you mentioned the issue of quality is one that you are looking at HETUS standards, but the notion of quality transactions, or whatever that might be, is something that might be in the future, looked at in the future.
From my perspective, I think about that in the context of how much a company like yours wants to be able to look at the information and extract it and do analyses of issues related to quality and performance of its providers and payers or its various groups, independently.
Have you given much thought to the availability and use of this information for secondary analysis, for management and performance reviews of your providers, or even identification of quality improvements in delivery mechanisms that you might develop independently of your providers or in conjunction with your providers?
MS. KOSS: We definitely do use existing information, as I mentioned before. They are based on the specified transactions.
Obviously we get our own kinds of cost reports, in addition to the kinds of HETUS measurements that, together, look at quality/effectiveness of programs.
We base our plan decisions on whether they will continue to be part of our program, or will hold enrollment current or release of plan, although historically we haven't done very much of that.
I have to say that our benefits staff are not RIT staff. So, they are not as -- they know what standards they are involved with, but that is kind of the limit.
We are not sitting there looking at exactly what could drive those transactions to better feed that quality effort.
We do know that, for what measurements we are using, we are very hard pressed to get consistent information, even as simple as some of the basic cost reporting within a plan on -- I don't have a good example of measurement off the top of my head.
DR. MORE: Any example does. Let me pursue that, if I may, for just a moment. If you are relying on HETUS-type information and cost reporting information, that is aggregated data from the plan.
You don't have access to the information to be able to aggregate it in a uniform manner yourself.
Is the availability of an administrative simplification so that all events and transactions have some standard approach desirable from that perspective, from your company's perspective?
MS. KOSS: I would think that, as a rule, we will continue to expect and require the plans to do it, that IBM, even as a major purchaser, is not going to get in the business of doing its own analysis, just as we encourage our customers to out-source their IT, because that is not their expertise. Analysis of the health care data is not IBM's expertise.
So, we would, instead, say, use these models. Use these processes. Use these standardized data to drive this analysis.
Then maybe we would engage either one of our third party administrators or even another outside expert to audit what they are doing to assure that it is consistent across plans.
In the coalition we do send groups of people on site to look at how they are doing, what they are doing to comply with our requirements in terms of quality reporting, and do on site visits. Again, it is leveraging an outside expertise that we don't consider our expertise.
MR. SCANLON: Shannah, IBM is an international company with facilities all around the country and other countries as well.
Has there been any experience in terms of state law or practices relating to standards that do or don't exist, or even international, that would be helpful as the United States considers standards here?
MS. KOSS: I have not heard about anything internationally or even on a state-by-state basis that has really risen to the level of this, or to say that this is a horrible kind of system.
I do know that on the business side of the health industry, that for the most part, although Europe is really -- there is the whole international standards driving it, that the systems themselves don't tend to be as evolved in terms of really network systems across all the care delivery.
So, you know, I have not heard about anything that would assist this, but I can go see if there is anything I have.
MR. SCANLON: I have a similar question about security. Have you come across any particularly exemplary approaches to protecting the security of this electronic information?
MS. KOSS: I want to get back to you. I mean, again, I am not aware, but I will look into whether there are some.
DR. LUMPKIN: Thank you very much. At this point we are going to break for lunch and we will reconvene at 1:30.
(Whereupon, at 12:30 p.m., the meeting was recessed, to reconvene at 1:30 p.m., that same day.)
A F T E R N O O N S E S S I O N (1:35 p.m.)
DR. LUMPKIN: Good afternoon. We are going to start off with our first panel. I would like to remind the members of the panel that, as you can see, there are sometimes some very pointed and detailed questions from the committee.
I would remind you that we did ask you for about a five to seven-minute presentation to get us started, and then we will go from the questions.
We will start off with Mark Segal. Welcome.
DR. SEGAL: Good afternoon. Thank you. I am Mark Segal, vice president of strategic market programs at the American Medical Association. It is a real pleasure to appear before you this afternoon.
The AMA is a very strong supporter of administrative simplification and looks forward to working with the NCVHS to effect appropriate implementation of the administrative simplification subtitle of PL104-191.
My brief statement is focused on answering the four questions that were posed in your invitation.
With respect to the first question, the AMA has very high expectations for this legislation. We believe it strikes an excellent balance between federal mandates and market development.
By increasing standardization and focusing on payer readiness to engage in standard electronic transactions, this new law will increase physician participation in electronic commerce.
This increased use of electronic financial and administrative transactions should increase efficiency in physicians' practices in the health care systems.
A lack of national standards has heretofore been a major bar to increased physician use of electronic data interchange.
By specifying standards -- and perhaps most importantly, by standardizing these standards -- the law will make EDI more attractive and useful to physicians, accelerating the annual increases that we have already seen in their use of EDI.
Now, increased specificity on technical and legal aspects of both privacy and confidentiality will also be very helpful in this regard. We will be testifying on privacy issues at the February 18 meeting of the committee's subcommittee on privacy and confidentiality.
Now, our major concerns with the administrative simplification legislation lie in the possibility that its implementation will overreach its core focus on administrative and financial transactions, to seek premature standardization of clinical transactions and computerized patient records.
For example, coding issues that are more directly related to CPRs than to administrative and financial transactions should not drive the coding standards used for administrative simplification.
Disrupting existing patterns of use of procedure and diagnosis codes, which are already well standardized, will only hinder administrative simplifications.
Similarly, there should be recognized that standards for the nine specified transactions have been completed by the ANSI ASC X12-N committee.
At the same time, of course, we support efforts to ensure coordination of pertinent standards developed by other standards organizations.
We are also quite concerned that implementations could, under some interpretations of the law, penalize payers and providers from mutually agreeable departures from specified standards.
A reasonable reading of the law, and one that is consistent with its intent, is that the specified categories of payers must be willing and able to accept standard electronic transactions.
This creates a proper expectation for physicians, that if they submit a specific electronic transaction that is standard with respect to EDI format and data content, that it will be accepted, and that if they do not, that their claim may be rejected.
A stricter interpretation, that physicians must submit only standard transactions to payers directly or through clearinghouses, and that payers can only accept standard transactions in all of their specificity, we believe that this interpretation goes beyond the intent of the law and would interfere in private transactions to an unjustified extent.
Finally, here, a specification of data content standards for non-institutional claims and encounters, the area that is of most concern to the AMA, should, in our view, adhere to the recommendations of the National Uniform Claims Committee, which the AMA is pleased to chair, and on whose behalf I testified before the full committee last November.
With respect to your second question, we do not have concerns with how the department is implementing administrative simplification.
To the contrary, we have been quite impressed with the skill and the professionalism of the staff from HCFA, from the Office of the Assistant Secretary for Planning and Evaluation and the staff involved with the NCVHS.
Now, on your third question, to date our members have expressed the greatest interest in standardization of claims and encounters, attachments and claims advice.
For claims and encounters, practicing physicians have indicated a great need for standard message standards and data content to ease administrative burdens and to reduce software, programming and transmission costs.
Many physicians have also expressed concerns about the transaction costs imposed by payers, networks and clearinghouses.
In response, AMA policy calls for many of these costs to be borne by the payers for whom the savings from EDI are so large and so certain.
We also believe the system ought to encourage more direct physician-to-payer transactions, rather than needing to go through a clearinghouse, for example.
Physicians have also expressed great concern with the burden of sending paper attachments associated with electronic claims, and the extent to which attachment requirements run counter to the push for automation.
From the physician's perspective, it is essential to standardize what must be in a particular attachment, when that attachment must be sent, and how the attachment may be transmitted electronically, rather than by paper.
It is also critical, however, to minimize the need for attachments altogether.
Now, standardization of payment and remittance advice, and automation of claims status information are also a major express need of practicing physicians.
A lack of consistent formats and information in this area continues to be a major source of frustration and costs for many practices.
With the growth of managed care, most of the remaining transactions will also simplify life greatly for physicians, especially increased automation of information on eligibility for services and for referral authorization.
Now, as providers of data, physicians face a mixed situation with respect to data availability. We share with you all an interest in collecting and maintaining data at the most appropriate site.
Thus, in general, we would like to see a reduced physician responsibility for data elements that should be collected and maintained by employers and payers; for example, certain non-clinical patient and insured patient information.
With the availability of the CPT and ICD coding systems, physicians do have the ability to convey to payers and others information on what was done for patients and why.
At the same time, we believe that improvements need to be made to paper and electronic data standards to allow for assignment as deemed appropriate by the physicians of at least four diagnosis codes for a given service. I believe that is a general view that the national committee has taken as well.
In addition, it is essential that as part of standards definition it be made clear the obligation of payers and others to accept and use CPT modifiers, which increasingly provide critical information on the nature of the service provided.
Now with respect to your final question, we emphasize that appropriate implementation of this law requires careful implementation with all the affected parties.
From all appearances, the department well recognizes this fact and is proceeding accordingly.
In general, we believe that the X12-N process to develop electronic financial and administrative transaction standards is appropriate as a means of meeting the intent of this act.
The AMA is a member of X12-N and, through WETI and the NUCC, also works closely with its members.
We understand that the department is considering using X12 standards for all but the professional claims and encounter standards. I realize that all of this is under consideration, but we have had a sense that that is the direction that things are going.
For the latter transaction, the department is considering seriously specifying the national standard format for the professional claims and encounters standard.
The AMA has taken no position on the relative merits of either standard in its present form at this time. Our policy supports both and, more generally, is highly supportive of the use of X12-N standards.
If the decision is made, however, to use the NSF as the initial standard for the professional claims based on a careful assessment of the relative readiness for implementation of the available standards, we do urge that a specific time table be set for a move toward the applicable X12-N standard, the so-called 837, with any needed refinements made as appropriate.
Finally, as the body that maintains the CPT code system, the AMA wishes to underscore a strong belief that, with respect to the reporting of professional services in the context of the nine specified transactions, that CPT can and will fully meet the intent of the law to "select code sets from among those that have been developed by private and public entities" with "efficient and low cost procedures for distribution."
Given the requirement that any standard adopted by the Secretary "must be consistent with the objective of reducing the administrative costs of providing and paying for health care," we emphasize the substantial cost that would be incurred by providers and payers if systems other than CPT and ICD for diagnosis were selected in the context of these standards.
With that, I thank you very much for the opportunity to participate this afternoon, and when the panel is done I would be happy to answer any questions.
DR. RICH: Thank you very much. My name is Ken Rich. I am a practicing dentist in a rural community of Kentucky and a member of the American Dental Association's Council on Governmental Affairs.
I believe I have submitted a more detailed written statement for the record and I think that has probably been distributed to you folks.
The first question the subcommittee asked the ADA to address concerns the association's expectations regarding administrative simplification standards.
The American National Standards Institute, or ANSI, is recognized by many standards developers as a central body responsible for the identification of a single consistent set of voluntary standards.
In fact, the health insurance portability and accountability act identifies ANSI as a starting point for the department's search for appropriate standards.
The ADA very strongly supports the adoption of ANSI ASC X12 formats, but suggests a migration toward those standards to permit the dental industry a reasonable adjustment period.
Currently a single dental EDI transaction generated from a dental office is normally handled by three different entities; a practice management vendor, a clearinghouse, and a third party payer.
The vendors market and install the EDI services and determine which clearinghouse will be used. The clearinghouses, which are few in number, focus on building relationships with practice management vendors and do not directly market to dentists.
The payers, their loyalties are divided among selected clearinghouses, generally based upon their ability to secure the EDI services for the lowest possible charge. No single clearinghouse serves all payers.
Most of the electronic transactions are currently being transmitted using a proprietary format.
As such, the ADA believes requiring all health plans, clearinghouses, and providers to immediately comply to a uniform standard would create a counter-productive and unnecessary upheaval in the dental community.
However, the ASC X12 interactive claim standard will soon be finalized and, when approved, should be the preferred standard to be used for the dental claims transactions.
The effects of proper implementations of standards to address administrative simplification should generally be positive for the dentists. Increased operational efficiencies are likely.
In addition, the ADA is pleased to note that steps are underway to address the means of facilitating provider reimbursement.
Currently only about 10 percent of dental claims a year are submitted electronically, but over 76 percent of the dentists have computers.
Of those who have computers, nearly 70 percent have the capability to attach or hook into an external data network.
The reason given for most of those who do not submit claims electronically include cost of submitting the claims, third party's inability to accept the claims electronically, and the burdens of dealing with the clearinghouse.
The subcommittee also asked if the association had any concerns about the agency's process.
To date, the ADA is comfortable with the opportunities to address NCVHS and work with the appropriate HCFA staff in a manner that enables the association to clearly convey its message.
In creating a process to achieve administrative simplification, the ADA believes that the Secretary should adopt the ANSI standards for specific transactions.
The ADA has actively participated in the ANSI's accredited standards committee for many years, and the ANSI ASC X12 standards that are being developed for electronic insurance transactions focus on the envelope used to transmit the data electronically.
The ADA has been working with the ASC X12 for the past 12 years in order to define a data content placed in that envelope, and is responsible for the data that pertains to dental claims.
The association looks forward to a continued dialogue as standards are established and updated.
Concerning the major problems that the ADA has with the current transactions specified under HIPAA, the ADA supports the adoption of standards for the financial and administrative transactions listed in the HIPAA, except with regard to the health claims attachments.
The ADA believes that attachments are not needed in dental transactions and only administrative data are necessary to process the claims.
Standardization of claims attachment information should be limited to a well-defined set of possible attachment, and eventually eliminated as a requirement.
Dental claims data, in its current form, does not provide researchers with the ability to study treatment patterns and the outcomes of care.
Only the service provided to the patient, and not the condition being treated, is reported to the dental claim.
The computer-based patient record and the terminology to capture the clinical detail and identifying characteristics are needed.
The ADA supports the development of a computer-based patient record. In fact, the association is actively participating in the development of such standards.
When electronic patient records standards are addressed, safeguards protecting the patient's privacy should be paramount.
Finally, the ADA suggests that the goal for administrative simplification can be achieved while meeting the business needs of all stakeholders in the following manner:
Thank you for your attention and I assume I will be on line for questions shortly.
DR. MILHOLLAND: Thank you. The American Nurses Association -- I did bring a copy. I don't know if you have a copy of this. I am sure you will get one eventually.
The American Nurses Association is pleased to have the opportunity to provide testimony to the subcommittee on health data needs, standards and security, regarding the administrative simplification component of the HIPAA.
I am Dr. Kathy Milholland, a senior policy fellow on staff at the ANA. I appear today on behalf of the ANA and its 53 state and territorial associations.
The ANA represents the nation's two million registered nurses, including staff nurses, nurse administrators, nurse scientists, and advanced practice registered nurses, including nurse practitioners, clinical nurse specialists, certified nurse midwives, and certified registered nurse anesthetists.
ANA supports the provisions of the HIPAA and our comments today are made in the spirit of assisting with the implementations of the administrative simplification requirements.
As the ANA understands this section of the HIPAA, the Secretary of Health and Human Services is required to adopt standards for specified transactions, to enable health information to be exchanged electronically and that, within 24 months of adoption of these standards, all health plans, health care clearinghouses and health care providers who choose to conduct these transactions electronically must comply with these standards.
We further understand that provider in this act means facilities, physicians, nurses, dentists, pharmacists and others involved in the delivery of health care.
This is an important distinction, since components of standards adopted by the Secretary will affect not only those business entities which directly conduct transactions -- such as filing a claim for payment for health services -- but also the individual practitioners who actually provide the health service -- for example, the registered nurse in the health clinic.
Thus, in this testimony, practitioner and provider are used interchangeably.
To respond to your questions, question one is about our expectations. We expect that implementation of these requirements will decrease the administrative paperwork burden on registered nurses, other health care providers, and health care consumers.
We expect that administrative processes will be streamlined. Duplication of administrative data collection will be reduced, if not eliminated.
Direct involvement of health care providers in processing administrative transactions will be reduced. The data collected and transmitted will be reliable, valid, secure and consistent.
It will be possible to aggregate the atomic level data from each consumer/provider encounter, so that valid comparisons between and among various groups can be made.
All health care providers, as well as health care consumers, will be uniquely identifiable.
The information in the transactions will be kept confidential, and the privacy of both providers and consumers will be protected.
One positive outcome of these achievements for our members will be to allow registered nurses to spend less time recording and processing administrative data, and reallocate that time to utilizing applicable information for the planning and delivery of patient care and/or the management of health care services.
This will, in turn, benefit the health care consumer through more informed nursing care and improved processes of health services delivery.
The adoption of standard code sets will enable nurses in different settings to understand the information being transmitted, reducing the confusion and the risk of errors.
However, it is critical that the code sets that are adopted include terms that describe the practice of nursing as well as the practice of other health care professions.
The inclusion of nursing terminology relevant to claims for the delivery of comprehensive health care services is essential. Otherwise, it is impossible to have comprehensive health care.
The absence of nursing terminology from adopted code sets will be a decided negative for nurses, for consumers, and for anyone who truly wants to understand the health care system in this country.
HIPAA calls for the Secretary to adopt standards providing for a standard unique health identifier for each individual, employer, health plan, and health care provider, for use in the health care system.
There are existing national provider identifier systems, as well as several publications available, that address issues around patient and individual practitioner identifiers.
The committee is encouraged to consider practitioner identifier systems which are already in use, instead of the creation of new systems or waiting for systems in development to be completed.
It is important that these identifiers be assigned to every health care practitioner who provides health care services, such as RNs, LPNs, LVNS, physicians, allied health workers, nursing assistants and so on.
For nurses, the assignment of a unique identifier will enable each nurse's contributions to patient outcomes to be identified and recognized.
It will strengthen the quality improvement process in all settings and promote the profession's commitment to accountability.
However, there are dangers inherent in the ability to identify individual providers and link them with specific patient encounters, if that information is used punitively, illegally or unethically.
Nurses and other health care providers will be concerned about the safeguards that must be built into the standards adopted by the Secretary.
At a minimum, as others have said, the identifiers must be designed without any inherent meaning to them. Strict regulations regarding protection of this information and severe consequences for violation for privacy are essential.
Furthermore, it is important that the wide spectrum of nursing specialties be addresses in the classification of health care providers.
Nurses' concerns over privacy and confidentiality of health care information extend beyond the unique identifiers to all the personal information transmitted electronically for both clinical and health care administrative purposes.
Numerous publications from various organizations address these concerns in detail, so I will not. I will note specifically that the code of ethics for nurses requires that nurses protect the privacy and confidentiality of any information about any health care consumer with whom they interact on a professional basis.
This applies to the collection and dissemination of any information via any media. Standards adopted by the Secretary must clearly and comprehensively address these issues, while assuring a balance between protection and necessary access.
However, the balance must always shift in favor of the health care consumer.
In regard to our concerns about the process, we support the process developed by the Department of Health and Human Services regarding the implementation of the administrative simplification requirements.
We believe that the opportunity for substantive public comment is very important, both to the department and to the stakeholders, who will be impacted by the Secretary's decisions. Thus, these hearings are very important and welcome.
ANA strongly supports the adoption of existing code sets, identifier systems, and transaction standards, with modifications as deemed necessary by the Secretary.
Our primary concern is that HHS staff may not be familiar with code sets that contain nursing terminology and transaction standards that support transmission of nursing information, or with the need to specify the inclusion of such information in whichever code sets and transaction standards are selected.
ANA will be pleased to meet with, consult and establish liaison with appropriate staff to assist them in the decision making process regarding these standards.
Question number three, when examining the problems experienced by registered nurses with the current transactions specified under the HIPAA, it is necessary to examine the problem from two perspectives, registered nurses as employees and registered nurses in independent or collaborative practice, which includes but is not limited to advanced practice registered nurses.
Registered nurses are employed in myriad settings, and their involvement with the data collected for specified transactions vary.
However, the RN in any clinical setting must collect a certain amount of the data used in these transactions.
Problems arise when this data collection becomes burdensome and interferes with the delivery of nursing care, especially when there is no visible return to the nurse for the effort to collect the data -- i.e., return on investment.
Furthermore, in most instances the current content of claims data is not clinically relevant; that is, it is not useful to nurses for clinical decision making or evaluation or care.
That is due in large part to the absence of nursing terminology in the data sets.
Advanced practice registered nurses and other RNs may practice independently or in a collaborative partnership with another health care provider.
These nurses experience many of the same data collection burdens as RN employees. In addition, they experience other difficulties in submitting claims and receiving payment.
Often, independently practicing nurses are required to obtain a pro forma sign off by a physician for work that the nurse has performed.
Another problem is the difficulty in obtaining recognition and payment for the nursing care components of their practice.
There are several factors involved in this problem. One factor, again, is the absence of nursing terminology in the code sets accepted by the payers.
This points up, again, the importance of ensuring that code sets adopted by the Secretary include nursing terms.
As we move from an illness oriented model to a more complex and comprehensive health care model, it will be increasingly important to recognize and pay for, or know the cost of, the contributions of nursing care to patient outcomes.
Question number four, we believe that these are not mutually exclusive issues. All health care systems stakeholders want administrative processes to be as efficient and painless as possible.
Adoption of existing standards with necessary modifications will be the fastest and most effective way of achieving this goal, rather than developing completely new standards.
Ensuring that the selected standards address the needs of all health care providers will go a long way toward meeting everyone's business needs.
The section on administrative simplification is a positive step forward in decreasing the paperwork burden associated with the management of claims and related information and increasing the responsiveness of administrative processes.
The adoption of national transaction standards will facilitate progress toward the adoption of comprehensive, accessible and useful electronic health records.
If we keep in mind that all systems, processes and standards are dynamic and can and will be changed, we can and will move forward without waiting for perfection.
Thank you again for this opportunity. I look forward to your questions and further discussions.
MR. JORDAN: Thank you for the opportunity to present to this panel the views of the professional pharmacy and the American Pharmaceutical Association.
I am Ron Jordan, a registered pharmacist, and president of H Caliber Consulting Corporation, also president elect designate of the American Pharmaceutical Association.
I have handed written comments to the panel and I am just going to basically summarize those, versus read them.
The American Pharmaceutical Association represents the third largest health profession, composed of more than 210,000 pharmacy practitioners, scientists and pharmacy students.
Since its founding in 1852, APHA has been a leader in the professional and scientific advancement of pharmacy, and in safeguarding the well being of the individual patient.
My testimony will focus on your questions. On balance, APHA believes that your work under administrative simplification holds great promise in both the short and long term.
As pharmacists, we believe we have a unique perspective on standards in the health care system, and that we understand how sometimes standards can sometimes facilitate, but also potentially hinder progress down a path toward improved patient care.
Pharmacists happen to be the most frequently encountered health care professional. They also rely more consistently on the use of computers in their practice than any other health professional.
In the United States, pharmacists in inpatient and outpatient settings rely on computers to perform their job exclusively.
The profession's reliance on automation and external electronic communication extends beyond simple communication of financial information and administrative information that is the focus of the HIPAA.
Pharmacists use computers to conduct electronic data interchange with payers, but also for patient record keeping, clinical decision support, and care delivery documentation purposes.
Pharmacists have extensive investments in their computer technology, and that computer technology has been based on the National Council of Prescription Drug program transaction models that have basically facilitated this extensive implementation of automation within the practice of pharmacy.
Any requirement for changes in the standards or the business model in use today in pharmacy will create extensive costs in an area where more than $1 billion claims were successfully supported in 1996 alone.
In spite of the costs and investment potential of your work, APHA recognizes that it is important to further advance and integrate pharmacy's electronic information exchange to improve the continuity and continuation of care, the quality of drug therapy outcomes, and patient knowledge on prescription drug use.
There are six key points that summarize our position on your first question.
We ask that the recommendations of the subcommittee preserve pharmacy's investment in our current electronic environment, requiring only those changes that are clearly justified and tested as to cost benefit.
We hope that the committee will endorse and encourage our profession's standard setting efforts and organizations, which continue to be effective forums for advancing our mutual agenda in health care.
We hope that, although the committee's focus is on the administrative and financial transactions within the health care system, we note that in one section of the charge of the committee, that you are charged with exploring and looking at the clinical area of health care record keeping.
We hope that your efforts in that area may enhance our ability as pharmacists to electronically connect with other medical professions, by focusing standards work going forward on required support for interoperability, rather than changes to existing standards that work for simply the fact of creating a new single syntax.
We hope that your recommendations as a subcommittee will relieve all health professionals of what we find in pharmacy to be severely disrupting and financial management burdens that result from a lack of progress to date in the development of unique identifier standards for patients, payers and providers.
We also hope that the subcommittee's recommendations will demand necessary and appropriate privacy and confidentiality protections affecting all areas of practice and organizations which make use of electronic health care data.
Through careful application of new security technologies, or encryption authentication and non-repudiation of information exchange, we believe that the most positive work the committee can have on our profession is in two areas:
First, in advancing the development of unique identifier systems for patients, payers and providers, also mandating the implementation of standards for patient health identification cards.
These areas will deliver enormous administrative simplification savings. We believe that in our profession this will also result in improvements in patient care, as the resources previously wasted on administrative and financial handling of these difficult areas is re-directed to caring for patients.
Second, we hope that your work will help accelerate standard development for the clinical data elements and objects that will facilitate interdisciplinary provider-to-provider electronic health care knowledge exchange.
We believe that, although there is a distinction between the clinical transactions and the computerized patient records that some members have spoken to, and the financial transactions that are the main focus of your charge, that they are interconnected in ways much like Kathy talked about, in that decisions made in this financing area need to consider the potential clinical standards implications in the computerized patient record.
On your second question related to the process being undertaken, we, too, agree that the committee's work on selecting standards and making recommendations to the Secretary are happening in an appropriate, open and participatory environment.
We happen to believe, because of our unique perspective in implementing, widely implementing, health care standards to handle the prescription drug area, that the perspectives of pharmacy may be useful to the committee.
You might consider more actively soliciting pharmacy's input to the actual committee work.
The major problems experienced by members, the transactions that are under HIPAA were apparently derived from the health financing-led ANSI ASC X12 business model for health care EDI.
Some transactions in the model advocated by ASC X12 are appropriate for pharmacy. Where appropriate, our industry segment has endorsed those X12 standards through NCPDP, and we are attempting to implement them.
You should, however, recommend that the Secretary encourage the continued use of pharmacies' NCPDP standards, in addition to X12, and other accredited standards associations transactions where appropriate, as I think has been mentioned by some previous panel members.
While we understand that the X12 batch model for health care delivery may work in the provision of other health care services, pharmacy has evolved past batch processing, separate eligibility and claim determinations, the use of claims attachments for COB and prior authorization, and the need for clearinghouses, et cetera.
Single, immediate response transactions in pharmacy support all these functions today, in addition to some limited clinical quality improvement functions that we are working on.
The specified transactions in HIPAA do little to support the clinical interface between providers. We acknowledge that even pharmacy's current standards are quite deficient from a provider-to-provider clinical information exchange and patient care optimization perspective.
Standards designs for our drug utilization review have not been adequately optimized yet by pharmacy payers, resulting in excess messages in pharmacy's EDI system.
This noise, coupled with frustration from a lack of unique identifiers threatens to limit our profession's ability to provide optimal patient care.
We hope that your work will encourage continued investment in improving these areas, clinical areas within pharmacy, rather than investing in a totally new administrative and financial transaction system for our profession.
Like other medical professions, standards and definitions of clinically focused health data elements and objects are needed to support the pharmacy component of the computerized medical record, and the desired automated, integrated health system.
APHA is participating in advancing pharmacy standards in these areas. APHA is also focusing its resources on advancement of pharmaceutical care.
It happens that in pharmacy an entirely new generation of information systems to support pharmacy practice are emerging.
These systems will support inter-disciplinary collaboration in improving patient drug therapy outcomes.
It is essential that the standards specified by the Secretary enable pharmacy's continued investment and progress in these systems.
Your recommendations must ensure pharmacy's compatibility and clinical information systems of other health care disciplines and organizations.
On how best administrative simplification can be achieved, administrative simplification will best be accomplished to identify the real problems that need to be addressed in broad based participatory process, further taking full advantage of the standards which are widely accepted and currently functioning in specific health care, so as not to impose undue administrative difficulties or wasteful expenses should be an important consideration.
Multiple syntaxes and standards can interoperate in the health care system, if appropriate business models, transaction data element compatibility and clearing requirements are specified.
Volatile market structures in the rapidly integrated health care system will ultimately determine the best standards, standard setting organization and business model.
However, it is clear that some areas of health care information exchange standardization will never be solved without stronger, higher power intervention.
For instance, in the area that we heard so much about this morning from HCFA, the area of unique identifiers requires fixing.
Pharmacy's experience shows that patient health identification cards are followed closely behind in this area.
Private side attempts to solve these problems have basically failed. We would welcome speedy recommendations by the committee to implement HCFA's new plans in both the national provider identifier and the payer ID, although we would implore HCFA to move forward faster and be certain that all patient health care professionals and payer organization needs are included in their plan.
Once unique identifier standards for patients, payers and providers are solved, the standards for health identification cards that are available, electronic signatures, practice locations and other functional implementation attributes need to be established.
I believe the coordination of standard development organization data dictionaries with assigned secretariats for various health sector domains should go a long way to ensuring success of your recommendations.
Your task it so build a standard structure for the health care system of the next millennium. I would define a system that makes it somewhat of a stretch for all stakeholders to achieve, and one that not only provides efficient and secure transactions for exchanging financial and administrative data, but also provides for the advanced architectures that were spoken about briefly this morning, which will be used to exchange meaningful clinical information between health care providers.
Ultimately, better coordination and exchange of information between health care professionals will lead to systems that we all desire for improvement of patient health care and optimal outcomes for the United States citizenry. Thank you.
DR. LUMPKIN: Thank you, all of you very much. At this time we will go to questions.
DR. IEZZONI: Well, on the old committee, I served on the medical classifications subcommittee. So, it was very interesting to hear two of the panelists -- Dr. Segal and Dr. Milholland -- talk a little bit about coding.
I guess we do need to address coding sets ultimately in what we are doing. The question I have for us first is, do we have to come up with a single procedure classification system? Is that a recommendation that we need to make? It is not. Okay.
DR. BRAITHWAITE: We don't have to come up with a single one.
DR. IEZZONI: In the past the committee has recommended that we move in that direction, and that may well be what we recommend now.
DR. LUMPKIN: So, we don't have to, but we could.
DR. COHN: Could I make a comment? I think we need to make some recommendation about procedure coding systems, about how they will be handled, what that is.
DR. IEZZONI: Okay, that is helpful background for my question. As we know, HCFA right now, Pat Brooks I see in the audience, is working with 3M to develop a brand-new procedure classification system.
The AMA continues to support the CPT. Does the American Nurse Association have a procedure classification scheme that they oversee and periodically update?
DR. MILHOLLAND: The association itself does not. The association has recognized three classification systems that have nursing intervention classifications within them that have been scientifically developed and tested reliable and valid that are in use.
That is the Nursing Interventions Classification, the Omaha System for Home Health Care, and Home Health Care Classification System.
All those have intervention codes that describe the nursing practice.
DR. IEZZONI: Does the American Dental Association, do your codes fit with the CPT, or how does that work?
DR. RICH: It is my understanding that we have our own set of codes which we have generated, and they come under CDT.
DR. IEZZONI: So, CDT is separate from CPT which is separate from some of the intervention coding systems for nursing and pharmacy?
DR. JORDAN: Yes. We are working on a coding system, too, for pharmacist services. NCPDP has an initial attempt at that for pharmacist services, and that is in the process of being expanded right now.
DR. IEZZONI: And Pat Brooks from HCFA, is the HCFA coding system, the procedure classification including nursing and dentistry and pharmacy?
MS. BROOKS: The answer is no. The new system isn't looking at expanding new types of categories that aren't here now.
So, the E and M type codes that are in CPT, we don't envision creating those, as we would not create a similar type coding system for the pharmacists and nursing.
We are trying to redo and update now what we have for use in the hospitals.
DR. STARFIELD: Can I just clarify? Do you mean for use in hospitals, or take what you are using in hospitals and expand it for outpatients.
MS. BROOKS: We use ICD-IX Volume III in hospitals now. What we are working on now is to improve that, to update it, for use in hospitals now. That is our focus.
DR. STARFIELD: Used in hospitals only.
MS. BROOKS: Yes, just as it is currently.
DR. IEZZONI: I think the committee is going to have to get a handle on this at some point. Probably now is not the time.
I think that we have had an eloquent expression by poor Pat, who I drafted, and our four panelists, that this is something that we are going to need to get a handle on.
The multiplicity of coding systems, that could frankly overlap. I mean, you have nurse anesthetists and nurse midwives which are perhaps the most clear overlapping situation with physician providers, for example.
I think, you know, that there is going to -- and nurse practitioners in the outpatient settings.
DR. MILHOLLAND: Even direct RN care, there are many things that both do. Then there are the different things that nurses do, which are not handled in CPT.
DR. IEZZONI: I just wanted to highlight the fact that we are going to need to talk about this further.
DR. LUMPKIN: Perhaps we could develop a coding system for each one of the coding systems.
DR. IEZZONI: A super-coding system.
DR. MOR: I have a question. Several of you mentioned concerns about the whole issue of attachments. In light of Lisa's question about the potential possibility of multiple coding, or activity coding systems which may or may not have direct relationship to the billing or the pricing of services, what do you envision as the notion of an attachment.
This is assuming that attachments might not be pieces of paper, but might be supplemental data strings.
DR. SEGAL: Why don't I start. I think there are any number of kinds of attachments that can be used. In Medicare and Medicaid there are, in effect, forms, specific forms to certify particular things that have a paper form and, increasingly, HCFA has developed electronic versions of those.
My understanding, from work the NUCC has done and others, is that in the non-governmental payer community it tends to be a little more free form.
It may involve X-rays. I think that is common in dentistry. It may involve extracts from the medical record, things of that sort, basically additional information that is requested to process the claim.
Again, one of the ways we are going and another kind of code set -- not really a procedure diagnosis set, but I think as you all know and will increasingly find, once you get electronic, there are code sets for basically everything, code sets for marital status and things of that sort.
We are very interested in exploring the use of either additional code sets or existing code sets, some of which exist, for example, in X12, to indicate X piece of information exists and is available in the physician's office. It is in the patient's record, in effect, attesting, under the overall umbrella of the physician or nurse practitioner, or whomever is signing the claim, that this piece of information, which they understand is necessary to help justify a service, does in fact exist, but not necessarily to have to actually transmit that.
Again, I think attachments can really cover the full range of additional data that payers or in a managed care environment, that may be necessary in addition to the initial claim.
DR. MOR: So, your testimony, then, was that such information not be requested routinely; that if it is necessary, that it could be supplementally requested as justification in that portion of claims or transactions that require clarification, but not be routinely requested.
DR. SEGAL: I think we would like to move to increasing specificity about in what circumstances attachments would be requested and then again, when they are requested, what form they would take.
DR. RICH: In dentistry, the codes for the most part are procedure codes and not diagnostic codes. So, we wind up sending in this information.
It could be anything from cast study models to X-rays to patient records to, in essence, justify the procedure that was done and determine the diagnosis.
To do that, that requires somebody to actually then go through that material and determine is this an appropriate thing.
I think part of the move would be toward some kind of diagnostic codes that we are going to have to deal with.
DR. MOR: That would be a totally different set of information.
DR. LUMPKIN: Does anybody else want to jump in on that one? No attached comments?
DR. COHN: I actually, first of all, appreciate Lisa's comments about coding systems. I agree with her, that would be a main focus of a subsequent set of hearings.
I also appreciate the speaker's comments about attachments. That was really one of my questions, although I still don't sense that we really have an answer about how to handle attachments from the comments.
I do, however, have another question which I had asked a previous panel, having to do with the issue of variability. Really, a number of you have alluded to needs for various things in standards as we approach them.
Earlier today we had talked about perhaps there would be a difference between the actual envelope standard and the actual data standard in that standard.
The question is, what kind of variability should we be willing to accept as we move forward with administrative simplification, versus what should we try to decrease the level of variability. I would appreciate your speakers comments on that.
DR. SEGAL: Maybe I will go first. I think in general we share what is probably the consensus view, that getting increased automation, getting increased use of electronic commerce is going to require considerably more standardization than is currently the case where I think it has been quoted that you have 400-plus proprietary EDI formats.
So, we very much support moving to standardization with respect to, in effect, the messaging standards, and with respect to how items are defined within that. You know, how do you define name, how do you define diagnosis, procedure, et cetera.
At the same time -- and it is a difficult issue and it is something that we have been grappling with, and Kathy knows, for example, in the NUCC, there are issues where, for certain payers, let's say, particular data elements aren't necessary.
This has come up, for example, with some state Medicaid programs, with respect to some of the providers they deal with.
To do what they are trying to do they don't need, let's say, certain patient information to be transmitted which, for example, we might include on the NUCC data set, because they already have that in their files. Or it is just information that may not be pertinent to the issue at hand.
So, a question has arisen in that instance. Need they require a provider to submit that if, in effect, they are only really requiring a subset of the standard transaction.
Should a payer, a physician, let's say, be penalized if, for whatever reason, some information is left off of a claim that is submitted, but the payer doesn't care because the payer, let's say, already had that information in their file, let's say information on the insured individual.
So, I guess what we would be looking at is the potential for some variability on which data elements are specifically required to be sent in every instance, in order for a standard or okay transaction to be considered to have happened.
In effect, physicians -- and we will be doing our part to communicate with them -- will know or should know that unless, let's say, all the data elements are submitted in a standard format, the claim may be rejected.
That is, in effect, I think from our standpoint, a sufficient penalty for not having complied.
On the other hand, in effect, there should be no failure to comply noted if some data elements are left off, ut the payer doesn't need those.
I think there we need to explore some degree of variability. Again, I think the other place to look for variability is at least initially in the implementation path where, for example, if HCFA goes along, or the department may consider let's say initially not using an X12 standard for one of the claims transactions, to ensure that in real live work, further improvement might be made on the relevant X12 transaction 837, that it could actually be used by willing participants, so that further development could go on so that, in fact, you could have a migration path to the use of that transaction.
So, those would be the two areas where we would be looking at potentially some allowance for variability.
DR. RICH: I think as is stated probably in the written statement, the dental claim form, or the dental EDI form or whatever is based on a paper claim, which we find some inadequacies in and some changes that need to be made.
I would not at this point want to sit here and enumerate those changes, but would be happy to get you that information as we have had time to take a better look at it. I think that is something that is ongoing and we are dealing with it right now. I would be happy to have staff get that for you.
DR. MILHOLLAND: I can't address any specific kind of claim, because I think in the area of advanced practice, the advanced practice nurses would have the same issues that Mark has pointed out, and the same desires for variability.
I think it also points to the burden of data collection to fill out a form, and the reduction of redundancy, if there are data already somewhere and it is non-changing.
Then it doesn't seem to make sense to have to keep sending it. I think that kind of tolerance needs to be established. That is where I would see some variability, too.
MR. JORDAN: Some of the things that I am hearing sound to me like what I refer to as the business model issue, and that the business model needs to be defined and, to some extent, standardized through the committee's work and the Secretary's requirement within whatever you can do inside that law.
It depends when you say how much variability we should allow. It depends on what level you are talking about in the system.
At the most granular level of data elements, I think you ought to permit no variability. I think that we ought to have uniform data elements for objects that we can define within health care.
There are certainly enough standards organizations out there with data dictionaries that could get together and form a single data dictionary without too much of a problem.
If we start there, we will be doing something in terms of ultimately getting toward interoperability.
At another level, I would hope that we even get a little more specific and define objects within health care that we are all going to need to do administrative work as well as do clinical work, and come to some agreement in that area, and that your activities kick into motion some work on defining higher level objects than data elements.
As someone said this morning, I think we might be trying to build a standards system that is going to be totally obsolete by the time we get it in place.
I am convinced that higher level technologies that are immediately available today could solve much of this problem, if someone would bite the bullet and say, we are going to force people to go a little further toward those other technologies that are out there.
When it comes to some of the things that people talk about on the burdens of who should accept data that is sent to them and how much should be required, we went through those things in pharmacy as we implemented the various versions that we had.
What NCPDP did was define some terms in the business model, basically, that had to be voluntarily adhered to, but we could put some push behind them by the way that we support the standards that are out there.
Basically, in pharmacy it is a central requirement, and it works that way in the industry, that when something is sent from a provider to payer, it is the payer's burden to take out what they want to use and throw away the rest.
I mean, it is not up to the pharmacy to necessarily maintain what each and every payer wants to see in our standard transaction, which might contain hundreds of data elements or as few as 50 or something.
Our standard was designed to permit providers that wanted to, to send as much or as little of that as their systems could handle.
So, some systems don't have that sophisticated a product and they basically send every single data element to everyone.
Others have more sophisticated products that let the provider decide how much they are going to send to each person.
There are definitions of where the burdens need to lie that we have worked through in pharmacy and I think ultimately everybody is going to need to do the same thing. Maybe the Secretary can do some of that. I am not certain about what latitude she will have.
Like I said, ultimately I think better technologies are going to reduce the variability, if we get into use of HTML objects.
We might even solve some of the attachment problems that are out there that we heard about, although I am not sure that we will be able to send casts of our teeth.
DR. STARFIELD: I have a question for Mark. You talked about your concern, the AMA's concern, that what we do with regard to transactions should have no relevance to computerized or electronic patient records.
Let me just make an assumption, and you correct me if I am wrong. I assume you could not envisage a situation where what we have in transactions is not translatable, or is not compatible with what might occur in an electronic patient record.
DR. SEGAL: I agree with that.
DR. STARFIELD: If that is the case, then what is your concern?
DR. SEGAL: I think it is not a concern. I think our point of view is not that they ought not to be relevant or not taken into account, but rather that the issues involved and the state of the standards and, to an extent, the state of the technology are at somewhat different places.
I think in our view that, for example, we are much closer to having workable standards in place that can be implemented as national standards for electronic data interchange transactions, than where we are right now for all the pieces of the computer-based patient record.
I think to an extent that is recognized by the time lines in the Kassenbaum-Kennedy bill, in terms of a date by which I believe this committee needs to do a report on that.
Clearly there is a relationship. Clearly one doesn't want to do anything with respect to the administrative transactions that prevents one from doing the proper things with the computer-based patient record.
At the same time, and to quote a former HCFA administrator, who probably borrowed this from someone else, I think we don't want to let the perfect be the enemy of the good; that with the kind of time lines that are in place now, I think we can accomplish a substantial amount of cost savings and efficiency in the system, by focusing initially and primarily on the administrative transactions.
We would not want to, in so doing, unduly disrupt existing relationships, again in terms of the huge investment that has been made by payers, providers, coders, et cetera.
Just to use the example of procedure and diagnosis codes, by trying to make the decisions now, and then sort of -- I don't know exactly what the right word is -- fitting them onto the administrative transactions in terms of the kind of coding systems that might, of necessity, be necessary.
I know Simon has worked heavily with this, with the computer-based patient record.
So, it is not that they should have no relationship. They should. Our key focus now is getting this first part of Kassenbaum-Kennedy done and up and running.
DR. COHN: Ron Jordan, I actually have a specific question for you, just following up about the variability question, and this has to do with the question of the impact of a pharmacist on a world of some potential variability.
What would be the impact on pharmacists in general if it was decided by the Secretary of HHS that NCPDP standards were appropriate for the role of pharmacy, but that everything beyond that was a different standard, such as information coming from physicians or dentists to pharmacists in electronic format, prescription orders and information going back. What sort of burden would that be for a pharmacy?
MR. JORDAN: I think that is what I was referring to as interoperability, Simon. I believe that we can build systems that can handle that bridge between pharmacy world, internal world if you want to call it that, and other areas.
It is probably -- I didn't mention that if you wanted to go at the level of packaging, like you talked about, of envelope and content, there is probably a way to package, in a way that is acceptable, an object that is an NCPDP claim, that ultimately ends up being translated by a batch translation system further down in the system where a payer needs to create a total multi-service longitudinal profile of what is going on with patients cost-wise.
I think as long as that burden of where the translation is placed outside of our world, the burden won't be that big a deal.
Expecting that pharmacy will be able to handle the structures and revamp its current infrastructure to handle X12 syntax, for example, or HL-7 syntax, is probably unrealistic, that there just isn't enough resource there. We are having enough trouble evolving our own internal world.
If we all of a sudden have to focus, strictly for an administrative purpose, to connect a little better to the world to go into a new syntax, that would be a significant burden.
I am not convinced that we can't have these clearinghouse bridges, these translators that go syntax to syntax.
We are able to coordinate right now with HL-7, for instance, with the prescription transaction that NCPDP is working on.
We don't think that translation back and forth between our two data dictionaries is that difficult. It is just a matter of sitting down and saying, these data elements are the same things. Let's make sure that they are the same size, they contain the same attributes, and that they are maintained the same going forward.
DR. LUMPKIN: I kind of have a general question, as seems to be my trend today. I am not going to necessarily hold you to answer it right today, but certainly to submit in writing later.
It is a term that is not often used in this kind of committee and these kinds of discussions, but it is a term called vision. Where is it that we want to go. That is not something that I have heard today.
We are dealing with a section of the law that talks about administrative simplification. Other than to mention that, I haven't heard anything that tells me how we as a committee will know that what we have done has met that goal. You can standardize without simplifying. We can just standardize the burden.
There are many practitioners right now who -- I would like to have some feeling that 18 months from now or two years from now or three years from now, that we have actually made something better for somebody who is providing clinical care.
To that extent, I would like to charge you -- and the issue of attachments is a classic one. We could standardize attachments, but we haven't addressed the issue of how much is legitimately in question.
At one time someone had one case where they needed "more information on" and now they are going to ask it on every single case, that they want to have more information.
I suspect that many of your members will have those kinds of details, those kinds of things, those kinds of stories that we need to understand so that, when we are done with our recommendations, we have some test to say, yes, we have actually made a change; we have actually made a difference.
It is too easy, because too many of us get caught up in the whole issue of standardizations and code sets to lose sight of the whole big picture.
So, I would like to encourage you, not necessarily today, but to think about that and submit something back to us, that allows us to keep touch with the vision of what administrative simplification is all about.
DR. MILHOLLAND: If I can just say, I attempted in some way to address that in what we expected to see as a result of administrative simplification, in that there is less involvement of the practitioner in putting together this information.
Sometimes there is a benefit in being rushed through having to put this stuff together quickly. Then when you start having time to look at it to see where you might expand it, I believe that it links back to what we call atomic level data; that is, the data that is collected between the practitioner and the consumer that is then placed into an electronic record. Then from that, devolves everything else.
That administrative data falls out of that without anyone having to do anything other than request the report. The codes are attached to it. The person uses natural language, but there is a translation that goes on behind the scenes.
So, the practitioner and the consumer have their interaction. The necessary clinical data are collected along with, at some point, the necessary demographic data and it is handled.
That is the simplification that we are seeking, is that we don't have to be involved so much in taking that information and extracting ourselves and converting it or having a staff of some humongous number being specialists in translating this information into a claim, or into some record of the cost of care. To me, that is what the simplification will do.
MR. JORDAN: Just in that area, I wouldn't want the consumer, the woman who spoke first -- Norma -- I wouldn't want her job in dealing with the claims for all the money in the world, because of the complications today that are being experienced.
In pharmacy, you know, the complications -- and we have a very simple little area that is fairly well standardized, but we still have enormous problems.
This unique identifier area, right now, in January, pharmacists all over the country as we speak are trying to figure out all these new cards that were just issued and where to send the transactions and who is really covering it. It is really unbelievable.
Basically we are driving as hard as we can to get you to get that decision made soon, so that at least everybody can be lined up to use the appropriate payer, patient, provider identification systems.
As soon as you can announce that the HCFA stuff is going to be it, I think the world will begin to align.
DR. SEGAL: It seems that this area, unlike a lot of government programs, really actually ought to be fairly straightforward to do a formal evaluation on.
Probably some of the specifications and some of the baseline data that could be used for an evaluation ought to be collected.
I am not sure if that is part of your charge, but perhaps that is something you could consider.
We could look at statistics on the amount of claims and encounter information, for example, that is transmitted electronically. There are people who collect that.
We could do studies on things like, well, is using standard transactions, does that actually save money. I think that is an empirical question.
We have an assumption that moving from 400 separate transactions to one will increase the use of EDI and save money.
That is testable. I mean, I think that is something that can be evaluated.
We can do satisfaction studies, and probably ought to be. How do nurses, how do physicians, how do office staff, how do they feel about the environment in which they work, in which they deliver health care, as a result of some of these changes.
Can more time be spent on patient care and less on administration. Or as someone said now on attachments, are we now, because we can do it, are we going to do it in every instance.
I think this is one in which people have assumptions, many of which they have made explicit, that lay out a vision.
I think we have a real opportunity to sort of test the intervention and to define some measures now and to maybe try to commission some studies, put together some evaluation plans, so that we can see, again, about whether it has worked and hasn't and where, and what kinds of corrections are necessary.
We would certainly be happy to follow up and think more on the vision. I think this is one that I think this committee could play a major role in.
MS. GREENBERG: I have a specific question that I hate to ask after the vision thing.
MS. BORDER: I would like to make a suggestion to the committee, that they give an assignment to the panelists who have appeared today.
When we talk about standards and codes, it is kind of like all of us describing the elephant from different points of view.
I think the committee should ask each one of the entities or representatives to give their pie in the sky vision of what a patient encounter should consist of, say, in the year 2000.
What information would be immediately available, what would the source of that information, and how would it be supported.
If you ask those of us who have already spoken to you today, you would find that there were overlapping circles and assumptions of where we think we are going to get this information from and how that availability is going to be the bottom line result of administrative simplification.
Once we find those overlaps, we may find out there is a lot more consensus out there than we may even be aware of at this point in time.
So, ask us what the vision is of what we think the ideal patient encounter would be for our constituency, and describe it to you and then support where the information should come from, because we think this is going to cure it all. Then see how much overlap we get.
DR. LUMPKIN: Thank you. I think that gives some structure to my earlier request.
DR. STARFIELD: Can I follow up on that? You are aware, I am sure, of the core data elements the committee came up with. Do you see a connection between your question and that, the core data elements report that came out this summer from the committee. Are you addressing the same sort of thing?
MS. BORDER: We know what the core data elements are, but how do they operate in the encounter. If I am a patient, what do I come in with. What minimal thing do I need to come in with.
If I come in out cold on the street as a result of a car accident or I am walking in the door for elective surgery, what moves my care along to the satisfaction of everybody involved, because the data is available about me and about the care I need, and that everything supports that.
DR. LUMPKIN: I think what we will do as a committee is perhaps, in some of our time on a conference call, we will frame this question and send it out to everyone on the current panel and on future panels.
I think that is an important question that we at least have some concept of where it is that we ultimately want to go, and we can test our progress to that.
MS. GREENBERG: I just want some clarification, Kathy, when you were talking about the unique provider identifier.
Were you recommending a system for the nursing profession different from the one that is being recommended under the NPI?
DR. MILHOLLAND: I was recommending that nurses and all other non-nurse practitioners be included. I was recommending that this committee and the committee as a whole needs to look at other identifier systems which are being implemented, as to whether what is being proposed by HCFA is what should be done.
I am not saying anything bad about HCFA. I think they have done a lot of hard work. But I have some concerns which I have frequently expressed in my reviews of their work, that they are still very claims oriented and they are not oriented to a bigger picture of a computer based health record wherein we can identify every practitioner and look at the quality and the cost and all that other stuff.
Their mind set from where they are coming is claims. I think the way they have structured their whole path, I think it is going to be a very long time before we get down to an individual practitioner level.
What I was trying to say was we have got to have everybody identified and not just the person who submits the claim.
We need to look at, is there at least one other system out there that is implementing practitioner level identification.
I get a sense, from the way we are talking, that the HCFA proposal is a done deal. That may be true. I am just saying that I recommend that the committee consider that other thing.
It is not a separate one; certainly not. Everyone must be under the same plan or we will be bananas. Being as nursing is the biggest group, we have a lot of interest in making sure it is a good system when it gets identified.
MR. JORDAN: We also would like to see acceleration of that area within HCFA. Even though I think NCPDP endorses HCFA's work, it is on the structure and the plan, and we would love to see it accelerated in some way.
Maybe that could happen if a model was defined where either HCFA or whatever the committee suggest be used is through a business model kind of brought together.
We might use multiple identifier systems to drive faster and accelerated expansion of whatever the ultimate system is going to be.
MR. MOORE: On the acceleration and the faster mode, when we did present the plan to the Deputy Secretary on doing all of this including the identifiers, he said we would do it with existing resources, both dollars and people.
We have drafted people from all over, everywhere, and we have asked for dollars. But there are constraints on us as well as anyone.
I think we are including nurses in the NPI. So, we are not ignoring them.
DR. MILHOLLAND: Yes, you are, but it is seen as a way down the road. As in every system I go with, it is seen as a way down the road.
My concern is only that there is still a very strong claims orientation and not a vision. I don't hear -- today was really the only time I heard a vision that this number might be used for something other than claims.
MR. MOORE: Is it the concept that we presented on an NPI what bothers you, or is it the schedule by which we are going to organize it.
DR. MILHOLLAND: I think it has for a long time been the concept -- and as I said, today was the first time that I heard any indication that this number might be used for other than claims processing.
That has been a serious concern because 2.2 million registered nurses, there are not that many of them that submit claims, but they are the ones that are collecting the information about patients in most settings.
They must be identified, they and all the others, if we are to truly understand what are the variables that influence patient outcomes.
MR. MOORE: We have two problems to solve. One is what I call the administrative simplification. That is, the electronic commerce that we are trying to wrestle with, and then the long-term view of how to improve the quality of health care.
I see one coming before the other, but I would hope that we are wise enough that we have the vision, as John said, to know where we need to be, and that we don't create something that would prohibit us from arriving at that point.
DR. MILHOLLAND: Precisely, and that has been my concern.
MR. MOORE: We can't do them all at one time because we don't have the resources.
DR. MILHOLLAND: I know, but if the vision that I keep hearing in the initial development of a system is a claims only oriented vision, that was my concern.
You have to start somewhere and you move with what resources you have. But if you are developing a system out of strictly a claims oriented process, you may not be able to extend it to the broader configuration. That is all. We could talk for hours about that.
DR. LUMPKIN: That is where we need the input, is that we certainly are looking at developing a system. There are many ways that we believe that medical and other kinds of clinical care can be improved.
I have a personal bent from public health, believing that the more data we can get on surveillance, the better.
I also try to temper that with the fact that that means people filling out more and more forms. So, somewhere we need to guide the system so that it has the ability, if we get to a computerized patient record that is acceptable, that is functional, and much of this other data that other individuals may want to extract is available so that the system doesn't lose granularity, does not lose data, but on the other hand is simplified enough so that there will be, in the year 2000 and the year 2005, it is my belief a significant portion of the system operating on paper.
Even if we conceptualize EDI, EDI is just how the data moves from one point to the other. It doesn't say whether or not I as a clinician have to write down the same bits of information two or three or four times in different ways.
So, we need to focus in, I think, on simplification. We need to also hear those comments, those visionaries who say, there are a couple of ways you can do it, but if you do it this way, then you can't do this down the road.
That is the bigger picture. That is certainly the kind of role that this committee has, where this committee is not a billing committee. That is not what the National Committee on Vital and Health Statistics is.
It has its roots in basic public health and statistical analysis of what is going on in setting directions. So, we are certainly very concerned about that.
MS. EMERSON: I am Mary Emerson and I work at HCFA and have been involved with the national provider identifier effort from the very beginning.
We never, from the beginning, saw it as a purely billing or claims related identifier. It was meant for enumeration of all health care professionals.
We, from the beginning, tried to develop a very elaborate taxonomy that could be used for other providers. I think Marjorie, who worked with us, can support that.
I am sorry that the message probably didn't get out, because there has been a lot of emphasis, as well, on billing. But it was meant to be broader than that.
MR. SCANLON: Again, just to go back to sort of an overview, the law, as you recall, spells out about nine specific transactions that are to be standardized for electronic commerce.
In your experience and judgement, obviously the ideal would be to be able to work on all nine simultaneously.
Are there, in your view, some standards that are, in terms of readiness and in terms of consensus, opportunities there to be implemented fairly earlier than others.
I guess our chair, Don Detmer, would refer to these as low hanging fruit because of their state of readiness and their state of use and the state of consensus in the field now.
In your view, what would the top three or four be, if there are such.
DR. STARFIELD: You might want to review what those nine are.
MR. SCANLON: These are the transactions: health claims and equivalent encounter information, health claims attachments which we discussed at length, transactions relating to enrollment and disenrollment in a health plan, eligibility for a health plan, health care payments and remittance advice, health care premium payments, first report of injury, and two more, health claim status and referral certification and authorization.
Are there perhaps two, three, four even that are well enough along to be focused on initially?
MR. JORDAN: I can make it a little easy in that pharmacy, in at least two particular cases, has adopted and is ready to move forward with enrollment and payment and remittance advice.
Clearly those are batch processes that are uniform, even in our heavily automated industry. Those two might be easy, although the design of the X12 payment and remittance advice involves the banking industry in a very complex way from a business model point of view. It might be harder than I make it out to be.
MR. SCANLON: In addition to the transaction standards, there are so-called supporting standards which deal with identifiers for providers, plans, employers and individuals as well as the code sets and classification systems.
So, if you threw those in as well, is the idea of the health provider identifier pretty well the first to be pursued?
MR. JORDAN: To me absolutely. I don't know how many privacy confidentiality advocates are here, but I actually believe that what we ought to start with is a single uniform identifier for each individual that is related to health care.
Then we can build the whole data base right off of that. Say I might be a provider sometimes, I might be a patient sometimes. I might be the CEO of a managed care organization sometimes.
If we could just get that done, when President Clinton held up that card a while back, I thought we were going to get there. I quit my job and became a consultant.
DR. MILHOLLAND: I agree with Ron. Actually -- I am going to throw out this question which doesn't answer your question but I know this committee has not discussed the individual identifier, which is probably the hottest issue that anyone has to touch.
I don't know if you are going to touch it.
DR. LUMPKIN: We are trying to stay away from it altogether.
MR. MOORE: I guess the answer to that question -- and Bob Gellman is not here -- but very much associated with that individual identifier, that is where we get into the real privacy protection, et cetera.
While I think that providers have a right to privacy, I don't think it is to the same level as the individual with the medical record. Payers have no privacy, evidently.
That is going to come, I think, later this spring as we develops for the privacy. I think we are on the hook to do that some time in the May/June time frame, with something going to the Secretary and Congress in August.
I think at that point we will be wrestling with what that individual identifier will be and it will be a little later this year. We are rushing it if we did it now, I think.
DR. BRAITHWAITE: We are letting the subcommittee on privacy and confidentiality take the heat.
DR. STARFIELD: But you are suggesting -- I think you are -- suggesting something different. You are suggesting, let's not have a unique provider identifier, let's not have a unique payer identifier. Let's just do it all together.
MR. JORDAN: Yes. Early work with HISPP on this issue said that if we could develop an identifier system that identifies both the individual and the organization, then we would have it all.
Then it is a matter of just saying what role are they playing in the health care system. Although it may make some privacy and confidentiality advocates cringe, I think there are going to be ways -- and certainly there is a start with the legislation that is being talked about now -- to protect patients' confidentiality and privacy in the systems.
I think actually existing technologies are available, if applied carefully, to do encryption and non-repudiation and authentication of the transactions, that can adequately protect the privacy of everyone that needs to be protected.
DR. STARFIELD: I guess the response before about what HCFA was trying to do is relevant in this regard. The comment, if I interpreted it correctly, was just to have an enumeration of providers, not necessarily to identify providers for patient care, associated with patient care.
So, I think your suggestion really gets at the heart of the way we are approaching it. I think we all differ on what the purpose of these unique identifiers is.
MR. MOORE: Well, it has taken HCFA three years to get to this point about the unique identifiers for providers.
I don't know how long it would take us to fight all the other battles if we throw all the other problems in one big bucket and go off fighting all the wars with those.
This is a HCFA point of view and mine, but I think we are, as a country, charged to get something done and not to continue to debate this forever.
We have been debating a unique identifier for individuals, I know, for more than a decade.
DR. LUMPKIN: We probably won't resolve that one today, because certainly we are moving down a path for separate identifiers. I am not sure we are ready to look at this unique one.
We always deal with things in caution. We had a bank and we had a joint account, my wife and I, and then I had a personal account, because I have trouble keeping a checkbook where somebody else writes checks.
DR. STARFIELD: That was atypical. Usually it is the other way around.
DR. LUMPKIN: What happened was the bank, even though we had two separate accounts, never could quite make the deposit into the right account, because my name was on both of them.
So, I am a little bit leery about our ability to handle unique identifier where my identifier when I am treating patients and my identifier when I am going to seek health care are the same.
There are all sorts of nightmares that can come to mind with that one. So, we are down that road and perhaps we will --
DR. MOR: I guess it is back to the code sets on the activities. I have been sort of stewing about this. One, there is nobody from the mental health world up on the podium now, and they have their own separate set of activities and diagnoses and so forth.
As I was thinking about it, I came back to the example that some nurse practitioners and physicians end up doing similar kinds of activities, and yet there would be different coding systems.
Just for a moment -- and some dentists do surgery that medical surgeons do. Clearly mental health professionals and physicians do the same thing, but then use totally different coding schemes for characterizing those behaviors, depending on who they think might pay for it or not pay for it.
If you sort of get an amalgam of all those things, could you imagine a co-existing set of activity codes that might be sort of coded to better characterize -- how can these co-exist on the same plane?
DR. MILHOLLAND: I can say that the nursing classification systems that the ANA recognizes and promotes, do not duplicate the CPT codes. They are the nursing interventions.
Where someone uses or performs an activity that is already captured by an existing recognized coding system, that is the one they would use.
So, the nursing intervention classification systems are uniquely to nursing practice.
DR. MOR: So, a physical therapist using the same kind of system would use a different classification?
DR. MILHOLLAND: No, they could use the nursing one, too. Why put them in -- we don't advocate duplicating stuff that is already there. We know that there are boundaries that intersect and are dynamic and that are fluid.
It is just that these people have done the work that is identified. This is what nurses do. This is nursing. But other people do some of these things, too.
Pain is a classic one. Everyone manages pain. Now, how you manage it may differ. But pain management is an area that is multidisciplinary. Some of the specific activities are different.
We don't see them as a big mish mash of the whole CPT code built into the nursing codes.
DR. IEZZONI: There are also ICD-IX codes for physical therapy. Those look pretty different than the CPT codes for physical therapy and they will look even more different -- Pat Brooks escaped -- when 3M is done with them.
DR. MOR: That is just for the inpatient side, not for the outpatient, which is the same procedure. It just happened to be done down the block.
DR. IEZZONI: Yes. The ICD-IX procedure codes are also used in outpatient care at hospitals. It is not just inpatient.
DR. LUMPKIN: I am a little bit surprised that Ken didn't jump at the issue of an amalgam of codes.
DR. SEGAL: Just from the standpoint of CPT and the others, I think there are a couple of issues. One, the CPT editorial process, recognizing that more and more health care professionals are using CPT for its various purposes, has increasingly involved non-physicians -- nurses, clinical psychologists, physical therapists, I can't recall who else -- in the advisory process that all specialties, for example, go through in proposing code changes and reviewing changes.
So, increasingly we have been working to proactively respond to the fact that there are multiple health care professions that are using CPT and that it is useful and appropriate for all of them, in situations where, for example, a nurse practitioner is providing a level three outpatient office visit, as opposed to specifically a nursing intervention.
I think the second issue in terms of whether we need the one coding system or whether we can have multiple coexisting is a recognition, I think, that is important, particularly if we are looking at electronic transactions.
My understanding of how, for example, the X12 transactions are designed is that there is what I think is called a qualifier which indicates which coding system is being used.
For example, if it is a procedure code, you indicate whether the procedure code that follows is a CPT code or an ICD with diagnosis or maybe whether it is ICD or DSM.
So, as long as you know which codes are being used, different people receiving that information -- some payer may say, we only want, for reimbursement purposes, CPT, or they may say something else.
But for the multiple people who are using this, I think you can accommodate whether it is for research or other purposes, collecting information that is not necessarily all from one single code set as long as, in the design of the standards, we are able to specify to some degree the universe of applicable code sets and have qualifiers so that both the sender and the recipient are communicating about not only what codes are used, but what coding system they are part of.
MS. GREENBERG: Obviously I know that is the basic pattern for the standards, that you separately indicate what system is being used.
But to really make that work in a vision of health care information system, wouldn't you need cross walks among them of some type?
DR. SEGAL: I think where there is overlap, where there is direct overlap. I think one would want to try to coordinate to minimize that. Where you have domains of nursing and general practice -- I think you want to avoid overlap.
I think where there is some degree of overlaps, I think you clearly would want to create cross walks. Sometimes overlap is going to be -- overlap could vary with the use.
There could be a researcher, for example, who in doing a particular study is interested in relating in some fashion, let's say with respect to the time involved in the service, a coding system.
Let's say you were doing some sort of research study where you wanted to take into account those services dentists did and physicians did.
One of the things you were interested in was how much time they were spending providing a service.
For that purpose, one might want to do some sort of cross walk comparing on the time spent, or what have you.
So, there may be reasons, even if code sets aren't overlapping, where you would want to try to do those kinds of cross walks.
In general I think you want to do real clarity. I think it is going to be, from a practical standpoint, you are going to have multiple code sets available.
I think you want to have as reasonably clean lines as possible between them, in terms of the domains that they are capturing.
DR. RICH: At this point, I think, in dentistry that represents a very small portion of what does go on. From that standpoint, I would question the need to make that a priority at this point.
I am just speaking as far as dentistry interacts with other professions.
DR. LUMPKIN: I think it is clear that coding is an issue that we will have to address later. I would like to thank the panel for coming and sharing your thoughts.
We have on our agenda to go to another panel. You may have on your agenda a break. Okay, 10-minute break.
(Brief recess.)
DR. LUMPKIN: The final panel for this afternoon, health care facilities and other providers. We have George Beatty from Mayo who is an add-on. See, where it says TBA, it means to be announced and we just announced him. George, would you like to go ahead?
Agenda Item: Panel on Health Care Facilities and Other Providers.
MR. ARGES: My name is George Arges. I am the director of health data management with the American Hospital Association.
On behalf of our 5,000 member hospitals, health systems, networks and other providers of care, I would like to thank you the committee for the opportunity to provide comments.
I also want to commend you for holding these hearings and for bringing together the views, perspectives and concerns of the different fields of health care.
Over the years, our members have had to deal with continual increases in the demand for data. The demand for this data is often diverse, complex and often open ended.
Our members collect and provide a variety of health data. Many data are used for medical review, billing, research and payment activities.
Users of this data include physicians, health care administrators, researchers, public health organizations, payers, employers, the public in general, data gathering organizations, and other providers of care.
Our members tell us that the diversity and the demand for data is costly in that it creates barriers to the full use of electronic communication.
We support the recent legislation that sets out to simplify many of health care's administrative and financial transactions by using standards.
The use of standards should help reduce our nation's expenditures for the administration of health care. We also recognize that the establishment of standards will also require changes in the existing system.
To achieve administrative simplification will require an examination of what, how, and when we collect data.
Ultimately we need to establish the role and responsibility of each participant to the process.
This would help providers to better understand their relationship to data content, format, and processing of information.
The causes for today's diverse data demands are a direct result of the deficiencies in the way the current financing of our health care system occurs.
These deficiencies, however, can be overcome if the process for data management is firmly established.
With respect to your first question in terms of what are our organization's expectations, our members expect a series of key reforms to be part of the administrative simplification regulations, and the uniformity and standardization of data content and format for each transaction set.
They expect clear expectations of the role each participant will undertake to collect, maintain and, if necessary, pass information from others.
They expect assignment of a data maintainer for each transaction. The process should include all relevant health care participants and include an open deliberation process.
Cost effect approaches to standards is expected. They must have stability in format design, their instructions can be easily carried out and understood, and efficiently transmitted.
We also expect that the process will establish rules that require all payers and providers to universally accept a transaction set once it is properly completed, without the need for additional attachments.
The transaction standard should fulfill its intended purposes.
We expect guidance as well on how conversion from current standards -- like the UB-92 format standards -- would occur.
Every effort should be made to keep changes to a minimum and, if necessary, follow an acceptable migration path to another standard.
Clearly the health insurance portability and accountability act directly affects each of our members, particularly with respect to the health care claim.
Our members have said that they do not want a costly conversion to another standard unless it is better than the one that they have been using.
They currently have attained a high level of success with the Medicare program, submitting approximately 97 percent of Medicare claims electronically.
They especially want all other payers to accept the UB-92 data set and format, as recommended by the National Uniform Billing Committee.
Our members also want to be clear about which data they need to routinely collect. This is needed not only to improve administrative simplification costs, but also to establish a basis for comparable data that could be used for research and quality improvements.
These results should give our nation a more effective health delivery system.
Your second question basically asked the concerns of our organization.
Specifically, we would like to see schedules about how the process will proceed. Is there a specific schedule on proposed rule making, along with the topics that will go out in the Federal Register.
Our members' concerns are centered on whether the Department of Health and Human Services would impose radically new reporting requirements upon the provider community.
Many other groups are seeking to impose changes and improvements ear marked mostly for clinical information systems. Yet, the law is clear in the stated purpose to deal with administrative and financial transactions associated with health care.
We, however, recognize the need to improve clinical information systems. The Congress has aimed the thrust of this law at the administrative and financial transactions.
The administrative and financial transactions mentioned in the law are important, not only for providers, but also for employers, payers, researchers, public health, business coalitions and others, who seek accountability of how health care dollars are spent.
One of the first decisions that the Secretary is likely to undertake with this act is the selection of electronic formats to carry the data content of each transaction standard.
The Department of Health and Human Services will be faced with a tough decision about whether health care will use ANSI ASC X12, or more conventional flat file designs like the EMC 4.1A.
While ANSI X12 holds a great deal of promise with their variable length format for each transaction set, their use in health care is very low.
On the other hand, flat files are more commonplace and used rather successfully by the institutional provider and third party payers in exchange of billing data. One only has to look at the success of the Medicare program.
Our members are worried about any change that may disrupt the flow of this data and the serious consequences it would impose on their operations.
Therefore, at this time our members are less likely to move to an untested and untried format like ANSI ASC X12 837 for the submission of billing data.
Clearly our members would not want to rely on any new format design that is not fully tested as a standard.
Adopted untested format designs could work contrary to the law, increasing administrative costs rather than decreasing them.
The review and selection process for any standard should be based on a rigorous set of criteria that evaluates the benefits and feasibility of adopting a new format as an industry standard.
The criteria should specify a series of key objectives that need to be fully satisfied before we adopt them as standards.
Once they have shown satisfactory compliance to the criteria, the nation can feel confident in setting the time table for migration to that new standard.
Finally, when the department addresses standards, it needs to be clearer about data content from that of format.
For instance, the UB-92 is a billing data set that can be reported into hard copy, the paper UB-92, HCFA 1450 form, or the equivalent EMC-4 format.
While the National Uniform Billing Committee designed the UB-92 data set to be a uniform bill, use of the data set, its content, has been less than uniform.
There are many -- too many, in fact -- inconsistencies among the payers as to their use of each data item, such as the type of bill or the type of service.
Compounding the problem are the inconsistencies among payers on being fully capable to use certain data items, like occurrence codes, value codes or condition codes.
These inconsistencies create problems for providers. The problems result when payers ask that they report this information that they normally would report on a UB on a different form or a different attachment.
The inability of some payers to handle all aspects of the UB-92 data set becomes problematic and administratively burdensome for providers.
Non-compliance causes providers to meet the payer demands through special programming, staff training and other manual interventions simply to receive payment.
We would ask that the department consider establishing a requirement on the universal use and acceptance of the data content, like the UB-92 data set, by all payers.
For the UB-92, it would mean that the NUBC items approved for billing would be allowed and any other item requested would not.
Providers would then need not report the same information on a different form, and unapproved items would disappear.
Finally, those payers who are interested in a particular data item for reporting purposes should have to go through the process, like the NUBC, to discuss the merits of their data request.
Today the NUBC evaluates the appropriateness of data requests and selects the proper method for handling and reporting the item.
Although content and format are two separate items, they are intrinsically linked together.
Care must be exercised in the selection of format standards. As was mentioned earlier, providers will need an incremental approach because it is the least disruptive.
To help providers better understand what will be expected from them, the department could help by establishing an end game for the future framework of how each health care participant will communicate in the future.
Your third question poses a series of questions looking at the problems experienced by our members.
As I mentioned before, our members find themselves developing customized routines to handle billing data for each payer.
Inconsistencies on what, when and how to collect and report data makes it difficult for providers to effectively train staff and develop appropriate information systems that accurately reflect patient services.
Complex coverage and reimbursement rules associated with health plans drives much of the special reporting today by providers.
Often payers never stop to analyze the rationale and the mechanics for providing this additional data. Consequently, these actions compromise the integrity of coding, particularly the clinical codes.
For example, many state Medicaid programs have developed special coverage and reimbursement programs. These programs often rely on the development of special procedure codes known as level three.
The level three codes are similar to CPT or HCPC but are not part of the official national procedure code list.
Another coding example in terms of inconsistencies, is the outpatient setting, where providers are often asked to report either the CPT or the ICD-IX procedure codes.
Selecting one code list would make the reporting easier for providers. In either example, providers report the codes the payers want, especially if they want to receive payment.
Many vendors that support hospital systems need to do special programming in order for the provider to handle these coding requirements.
When our members have to deal with multiple reporting requirements it makes it more difficult and costly in the long run.
As our health care system continues to consolidate, it becomes more difficult for providers to centralize administrative operations when they must be aware of these special codes, the local requirements, proprietary items.
Providers want a process for deciding which code should be used, and that process should be open.
It should be national in scope and, in the long run, the maintenance of these codes, the local codes, should be abolished because they cannot be justified.
Our members request that the regulations preempt the use of any local codes. This would force the issue to a national forum that could deliberate on the need for such codes.
With respect to the fourth question, how the goal of administrative simplification can best be achieved, again, our members want an incremental approach that will build from the existing infrastructure.
They are cautious of any attempts that seek to recreate the wheel. While there are inadequacies with the way the current process functions, we can make improvements.
Our members rely mostly on voluntary efforts of organizations. The AHA is proud to have played a major role in the establishment of the NUBC, the National Uniform Billing Committee, and its creation of sister committees at the state level.
The purpose of the NUBC is to promote the development of uniformity and standardization of the billing data set.
Again, our members look to the NUBC to define that billing data set. Although much of the work that has been done thus far has been voluntary, the NUBC has become, in many respects, a de facto standard with respect to the UB-92 for billing purposes in the institutional setting.
Also, it should be noted that other information systems within the hospital tend to work in concert with the UB-92 data set.
While AHA's efforts have accomplished much on a voluntary basis, we could have accomplished much more with the benefit of legislative support.
I am confident that those who have followed the deliberations of our committee would also find the transition to national standards easier.
Inversely, the organizations that have ignored many of the recommendations made by our committee are likely to find the transition far more difficult and costly.
Clearly, establishing the right framework to tackle the issues is important and some key points to remember are that the standards must apply to everyone, including federal preemption, that they include uniformity and standardization of data content and format, that they be cost effective in their use, that they rely on a balanced forum for deliberation, which includes all the major stakeholders, that the process itself be predictable in terms of its review and update, and that it utilize an effective communication infrastructure to make the information available to the users, so that data standards are understood and easily carried out.
I thank you for your time and I will answer questions after the other panelists speak.
DR. SCHULTZ: Thank you. Thank you for inviting me today. My name is Sam Schultz. I am vice president for information services of the University Health System Consortium.
Just to tell you a little bit about myself, I spent my entire life since high school in academic health centers.
I have been a student, a professor, a chairman, the University of Michigan. I have been a CIO at the University of Pennsylvania. Now I am an administrator at the University Health System Consortium.
By the way, we used to be University Hospital Consortium until about a year ago. Since most of our members now are becoming full fledged systems, we changed our name, but kept the initials.
I work with the CIOs of our organizations, the chief information officers, the people who try to move the data to its proper location, and all of the other Os who are involved in the administration of health care.
Let me tell you a little bit about the University Health System Consortium. By the way, there are some propaganda that hopefully has been distributed that describes our consortium.
We are a consortium of more than 70 of the nation's major academic health centers, and another 50 of their partners that comprise their health systems. As of last week we have formed an alliance with the NAPH, the National Association of Public Hospitals.
So, we now have members like Cook County Hospital, San Francisco General, Denver General, and those kinds of hospitals. So, our mission has widened a bit.
As you know, a good deal of our training goes on at those locations. So, there is a good reason for us to work with those hospitals.
Our base members consist of places like Georgetown here in Washington, Mass General, Stanford, Penn, Colorado, Alabama, Florida, Michigan and places like that.
We are a not-for-profit organization. We are 10 years old. We are a business organization. Our basic goal is to help our organizations become more competitive in their marketplace and to support their complex mission, research teaching, clinical care, in this ever-increasing competitive marketplace.
One of the most important functions that we provide is to collect data on their behalf -- clinical data, operational data, financial data, and to organize the data and provide it back to them in many forms so that they can improve the practice of health care and improve the operations and delivery of health care.
I am also, I must admit, the founding chair of HL-7, so I will also speak to HL-7 primarily today. I didn't get here when you began, but it sounds like most people here are familiar with HL-7.
I have included some documents so that if you are not, you can become more familiar. Basically, we are an ANSI accredited standards development organization.
We began about 10 years ago as a business organization to get together all parties to try to solve some of the problems in getting systems to talk to each other.
I think, as I say, many of you are aware of what we are doing today. Let me speak now to some of the questions that you asked.
Regarding the first question, what are your organization's expectations for administrative simplification standards in HIPAA.
If done correctly, the standards requirements of HIPAA could be greatly beneficial to members of our organizations. Anything which could increase the use of common emergent standards such as X12, HL-7, which for us have become more or less universal for the clinical enterprise of university health care systems, a financial or clinical data exchange would help greatly.
According to most of our surveys -- and we regularly look at all of the application systems that our members use, the specific vendors they use, what they are using it for, what data they exchange and what standards they use.
Something larger than 95 percent of our members use standards such as HL-7 and X12. Now those are only two of the standards. One of the things I included in the packet is a little inside cover reprint of a medical informatics journal that some of you may be familiar with that really lists message and code standards that really are the predominantly the suite of standards that we use for exchanging data and codes.
It is an ad for a place that we can go to as organizations to find out more about messages and code standards that Clem McDonald puts out.
So, we use many, many standards -- ICOM, NCPDP for example, for pharmacy. They are critical to our existence. They are critical for our improving health care and reducing the cost.
We also think that standards at the communications level would begin to embrace some of the best aspects of the internet.
I know we are talking about standards here at another level. But for us the use of the internet has become crucial in improving the quality of education and the ability to do research.
We would like to see that, whatever is done, continues that harvest, while at the same time attempting to maintain security and some of the problems that we have to deal with.
It is our expectation that this administrative simplification will result in some cost savings, by utilizing working and accepted standards in each domain.
For example, we are assuming that HCFA will continue to require support for UB-92 and 1500 formats, which we use widely in our data collection for research and clinical process improvement work in the consortium.
HL-7's version 2.2 standard, which supports these data sets, could be used for this requirement, clearly.
Besides offering support for these reports, HL-7 provides providers such as our hospitals and clinics with the ability to consolidate patient billing information between computer systems. We use it to do so, and the ability to transfer the components of medical records that we need.
As I said earlier, HL-7 is ANSI accredited, and it is now in widespread use internationally. That is an important thing for academic health centers because, as you may know, international commerce in health care is part of what we do.
Finally, on this first question, I think it is important to say that to exclude any standard that is now a significant part of the current suite of standards that our members use would, for us, create significant burdens. I think I am echoing much of what George said.
Second, do we have any concerns about the process undertaken by Health and Human Services.
Our members do have some concerns. Our biggest concern probably is one based on communication. That is, we have some question as to how fully DHHS understands the issues, and particularly the separation of administrative and clinical data specification and the difference between those as they are standardized, as the data is attached in reimbursement issues versus clinical, research and care issues.
We would like to see that HL-7 and other standard development organizations be given the opportunity to work directly with various departments to really develop an understanding at the level that we feel needs to be achieved.
We think HL-7 provides a robust offering in terms of clinical data. Clinical data is our forte, both inpatient and outpatient, unmatched by most standards. We want to assure that that strength is fully understood and embraced.
We do have some concerns about some of the privacy directions. I guess it sounds like that is going to be discussed later on in the year.
Let me just say that we would like to assure that there is a healthy balance between the protection of individual privacy and the need to continue to facilitate the flow of data for research and teaching. It is a key balance that we would like to see achieved.
Question three. What major problems are experienced by members of your organizations with current transactions specified under HIPAA.
I think we would echo some of the things you said, George. I think that is very much the case that our members have to deal with.
Particularly, again, the boundary issues. It seems that there is a good deal of mixing in terms of clinical data, with HCFA charge requirements. I think a better understanding with better definitions of clinical data within its own domain, properly separated from financial data is something we fully support and would like to see achieved.
If clinical data can be fully separated from financial data, as we attempt to do in HL-7, we think it could facilitate a lot of cooperative enterprises that we are already engaged in.
For example, cooperative clinical trials with NIH. The degree to which we can continue to achieve better standardization at a clinical level will allow us to do those kinds of things more effectively.
Right now we do a lot of borrowing of what I would call administrative standards. The separation of those two would help us to achieve some of those other goals.
Question four, how can the goal of the administrative simplification best be achieved.
It is our feeling this goal can best be achieved by utilizing a concert of existing and market proven standards.
We expect that a working and accepted standard will be selected for each domain.
Clearly, HL-7 is an evolutionary standard. We started out at the beginning assuming that we would have to make a lot of sacrifices, allow a lot of optionality so that we could bring people to the table, so that they would begin to discuss their needs as providers, their needs as software developers, their needs as users and so on.
What we are attempting to work toward is more and more specifications so that, in fact, standards can be certified, so that software can be certified, so that exchanges can be certified.
In terms of the earlier discussion I was listening to when I got here, we would like to push as hard as we can, and are attempting to do that, in the version 2.3 of 3.0 of HL-7, to achieve more and more specificity.
I think the key issue is the connectedness of the standards efforts to the underlying business process of the stakeholders.
If it is connected to a business process and everybody has that shared need within their business process, we think that it will succeed.
If there is a disconnectedness, our experience is that those standards development organizations do not succeed. We think that is a key issue. Thank you.
MR. RODE: Good afternoon. My name is Dan Rode. I am with the Health Care Financial Management Association, the group that gets confused with HCFA every once in a while.
I am director of the HFMA knowledge network. There is also a long explanation of what that does, so I will hold that for questions if you want.
I would like to thank you for the opportunity to testify. I would like to thank you for the work that you are going to be doing with the new public law, 194-191, and to continue to offer our services as an organization if we can help in any way.
HFMA is a 50-year-old professional association of more than 34,000 members.
Our members work across the country in various financial information and administrative settings. They work in hospitals, in clinics, in long-term care. They spend time in physicians' offices and home health agencies, and skilled nursing centers, and managed care, and some of them are also consultants as accountants, as information people and third party administrators.
So, they are kind of a gamut of administrative, financial and some information officers across the country on both sides of the payer/player table, so to speak.
Our members have been involved in the issue of administrative simplification for almost 30 years, which is kind of scary.
We were involved in founding the NUBC back in 1975, when we were certainly looking at hospital billing.
We were involved in setting up the health care task force of the X12, when the time came to look at transactions as they apply to all the settings, not just physician offices or hospitals.
We have also been involved with HCFA on the Medicare Technical Advisory Group over the last three or four years, to look at ways of improving the operations of that organization as it interrelates with the various players.
We were also involved in seeking the passage of the legislation that you are now dealing with. So, we feel some comfortableness in talking about that.
There are two terms I would like to just stress in my testimony. I am certainly not going to sit here and read all of it. You have a copy and you can look at it at your leisure if you get any leisure. But there are two topics that I would like to mention.
One of them is uniformity. It is our feeling that if we don't address the issue of uniformity, then the cost savings associated with administrative simplification will not happen.
We have got to deal with a system so that the information that we have seen in the two WETI reports, the information that came out of the Congressional Office of Technology, our own Lewen report, which I have attached for you, can show that we have to do business in one way.
Now in answer to a conversation earlier this afternoon, that one way may have some differences and some flexibility.
As we have heard from some other speakers, one piece of data means the same thing no matter where we are looking. A data set represents the same package of data, so that we don't have the inconsistencies that you have already heard about this afternoon, but can address those.
The second term that we need to stress is the term industry. The whole purpose of this law is to establish a uniform standard across the health care industry.
It doesn't apply to any one particular group of payers. It does not apply to any one particular group of providers.
It encompasses, unfortunately for all of us, well over a million providers, almost two million, I believe HCFA sometimes quotes, several thousand payers, a variety of vendors, and all sorts of people who work as consultants in and out along the way.
Let me address the questions you raised. You asked what was our highest priority and you also asked if our expectations were met with the act.
Let me say that most, if not all, of our expectations were met with the act as it was passed. We really stressed uniformity among providers and payers and essentially the act kind of puts the burden on payers.
As George has indicated, we would really welcome what this committee can do and the Secretary can do to make sure that everyone is responsible in a participatory way, as well as for the requirements to be uniform across the industry.
We were also concerned that the act allows states to opt out, that there is a possibility that the Secretary would give the states the ability to have exceptions.
I just point out that as soon as we have an exception we have no longer uniform standards. Patients cross state lines, data cross state lines. We can't have a change if we are dealing with someone across another border.
We didn't agree with the act's two-tiered implementation. Our feeling again is to implement at one point in time for large groups and another point in time for small groups means that we don't have uniform implementation until we get to that second point.
In the meantime, we have cost up front to the first group to keep going with the new system and the old system, and we also have the potential for other problems.
We really would like to stress and think that there is an ability to look at a single time. Back it up a little if you have to, but try to go with that. We would certainly support a change in legislation to see that occur.
We are concerned with the penalties that were placed in the act. We really think we should be looking at incentives and not penalties.
Our biggest fear with the penalties are that they are so low that actually you could get by a lot easier just getting the penalty than to participate in the whole process of uniform and administrative standards. We would like to see some benefits come out of this for everyone.
Finally, we have supported an industry-wide committee, or we originally said commission to work with the Secretary on this act.
We are pleased that you are part of that process at this time. We want to see, as George has mentioned and as Sam has mentioned, an industry-wide process, so that we can work with you and get the consensus we need to move these standards ahead.
We support it in the act and certainly support also as the committee goes forward, to look at the fact that we have the ability to use clearinghouses and VANs as ways of bringing providers who may not be up to speed, to use initial uniform transactions; that there are ways to do this without saying to every entity, you must comply with the standards.
We have ways of bridging between this, and I think it is important to understand that.
As other speakers have mentioned, we are also concerned that the committee look at the core transactions. We are aware that there are other transactions we would all like to see in place.
We are aware of the efforts on the clinical side, but we would like to start with the core transactions. We believe that these core transactions and a uniform basis could serve as a model to begin to build into other kinds of activities that I know this committee is interested in and I think probably everyone in this room is interested in.
The bill essentially gives us three years. We think that it is better to take the tone we heard earlier, remember what we are going for, but in the meantime the transactions need to follow the ones that are listed in the law itself.
Our members, what do they expect to get out of this? I think George kind of gave you some of that picture. Right now they are faced with a myriad of requirements. Whether they are a payer or a provider, they have got to figure out who they are dealing with, how to define what they are dealing with based on who, and then how to collect it, how to send it, how to store it and so on.
What you are looking for is a way out of that process. We are looking for a way to simplify the transactions that we are dealing with and to simplify the use and the collection of data.
Some believe that the vendors in our industry are going to kind of slow this down. That is not the case from our membership, who are vendors. Most of them welcome this.
A good example is the typical vendor who has to implement Medicaid standards. They have to implement 50 different Medicaid standards if they are lucky. They may have more than 50.
We can see the day when it will come down to one standard that we have to deal with.
When I was a finance director at the University of Minnesota and also at the University of Utah, I used to have specialists who had to deal with each major third party.
If I had a third party that was not a major third party, then I had to find somebody to figure out what the rules were, before we could even begin to transmit or receive data from that third party.
So, what we are looking for is a way to simplify that process.
To answer the question, it may mean not necessarily everybody using the same standards. But certainly between types of providers and third party payers that take into account pharmacies and physician offices and the rest.
You asked, which standard should we start with and that is difficult. We want all nine or ten, to be honest with you.
I think if we have to address them, it is most important right now to look at the 837, the claims transaction, and the remittance advice, the 835.
Right now we know that HCFA is already moving out on the MPS process. This is a major portion of that program.
We know of several states who have established mandates at the state level to begin dealing with this. Typically the area they worked on first had been claims and payments.
If we had the leisure of six years instead of your 18 months, we might start at the beginning, as someone else started, and begin with eligibility and enrollment and move our way back through the system.
But if we have to look at something first, that would be the area that we want to look at, and we think it is very important to, again, look at those from an industry-wide basis.
We are not looking at a system that is just going to support the MTS process, although we certainly want the MTS process to go forward. We need a system that will apply to all the payers and providers.
You asked about our concerns about the process being undertaken. I think our biggest concern right now is we are not sure what that process is.
We need to get more information out to the industry as to what is happening, how it is happening, and who is going to be involved in it.
We want to know what the rationale is for the decisions that are made. Most of all, we want to make sure that the committee has the benefit of a lot of years of work that have gone into setting the standards up that we already have and the needs of the process.
We recommend that the secretary and the committee put together a technical advisory process, and allow the experts to come in and deal with the peculiarities that you have already heard today, that just are really going to take a lot of time and a lot of activity.
It is no secret that the National Uniform Billing Committee has been around since 1975, and the National Uniform Claims Committee has certainly been working for two years.
Those two groups, along with the X12, have been working on common data definitions. That work can be added to the work of this committee, along with vendors who understand what it really takes to move this data through the systems, physician offices and hospitals and the like.
We really do have just a tremendous number of different groups that we have to look at.
We also suggest that when we begin to deal with this data and this uniformity, then we will begin to deal with the question of quality.
Right now it is very difficult to have quality data moving back and forth across our industry when we have different definitions.
When a clerk in an office can't remember which third party wants what and sends the wrong thing, as we have seen very many times the accusations of fraud sometimes are not so much fraud as they are, what was the rule that I was supposed to follow for that transmission of data.
How do we improve that kind of process and get everyone involved, so that we have that.
We also know that there is a significant learning curve. I have been working with EDI since 1989 and I know that my members don't fully understand it.
I also know that my members are frightened because the biggest fear right now is, if I go ahead and begin to move my shop to EDI, will the standards be there. I don't want to make any moves until I know everything is in place.
We have had a lot of reluctance, with the exception of groups like the Mayo, to move forward on this because no one was sure if anyone was going to follow.
It is real important, again, to look at the goal and where we want to move as an industry. I don't want to say build it and they will follow, but essentially a good part of it is going to be us knowing what these standards are, what you are doing, and begin to educate our members.
We have listed, on the last page of our testimony, I think what we know today and some things that we need to do.
We think we need to choose the transactions, again, for the entire industry. We need to be mindful of the overall goals and we need to be mindful that we need to interact with other industries in the United States, which is why we felt very comfortable with the X12 transactions.
Certainly from a clinical standpoint, eventually we have to look at international standards.
We feel that the work of the X12 is a good basis to work on. We feel we have to work with industry vendors. We need to implement standards and processes and new data sets where needed.
We need to highlight our commitment to the public and educate them. We need to jointly provide education and work for consensus.
We need to identify options so that those who can't make the move initially can move at some point in the future.
We all have a duty, I think at this point in time, to also pledge our work with you, and I wait for questions.
MR. BEATTY: My name is Gary Beatty. I am an electronic commerce specialist with Mayo Clinic. My area of responsibility is strategic planning as it relates to electronic commerce implementation.
When we look at Mayo Clinic, we are a large health care provider. If you look at our logo, we have three shields in our logo, one of which stands for patient care, one that deals with research, and one that deals with education, all of which revolve around quality care for the patient and what is important for that.
That is our main focus, not how can we bill patients, how can we handle all the various administrative tasks. So, we want to simplify that as much as possible.
We are looking to do that through initiatives in electronic commerce that encompass all the various aspects of standards and selection of standards, the data content, coding, identifiers and so forth.
Standard business practices, which I haven't heard today, is the issues outside of the standards, the data count. How do we put it all together to make it all work, so that we can live in an electronic commerce environment, and the various communication aspects that are available to be used with electronic commerce.
We have taken up those issues at Mayo, to implement dealing with clearinghouses, value added networks, doing direct connections, and also recently doing what we call secure EDI over the internet and intranets.
To support that, we are looking for this committee to address the administrative simplification requirements with a recommendation to the Secretary to adopt the standards from the American National Standards Institute, and Accredited Standards Committee X12.
X12 has developed transactions to meet each of the requirements outlined in section 1173-A, all nine of them, as well as meeting the data requirements outlined for coordination of benefits.
We have implemented many of these transactions. In my past, I came from a payer environment prior to this year.
In that environment we had implemented the 837 for professional claims, for institutional claims, and for dental claims in multiple versions of the standards.
We also implemented the 834 for health care enrollment, as part of our processing of the health insurance plan in California, with 35 training partners in a fast implementation process that we implemented within a 60-day time frame.
We implemented the 271 for eligibility roster and shared that within our pharmacy setting. We also implemented the 835 for health care claim payment.
Recently, we are in the stages of piloting the 277 claim status transaction with HCFA.
As we have gone through that implementation process, we have found that once we have done one or two, the process becomes easier and easier and easier.
Some of the benefits we are realizing as part of this process is that the cost of implementation has gone down, from implementing proprietary products like the National Standard Format, and the flat file for hospital claims.
We have also found that the implementation time or the turn-around time has been significantly reduced in the implementation.
When we were implementing the versions of NSF or the UB-92 flat file format, it was taking us on the average of about two months to implement a new training partner.
When we moved to the 837, for example, it was taking us a matter of five to ten days. So, it was a significantly reduced time frame.
All of that revolves around the process and business practices that we are looking to implement within all of our electronic commerce initiatives.
This will allow us to concentrate on our number one priority again, being patient care and high quality patient care.
We realize that there is a considerable effort in switching from proprietary formatted standards. We also have realized, through our experience in the implementation of the various X12 standards, that those costs will be far out-weighed by the benefits that we have already realized.
As far as the second question as far as any concerns with the process, we think that the process is very good that you have in place.
I would like to reiterate one statement that Dan had mentioned, that there are states out there that have passed legislation dealing with health care simplification, such as the state of Minnesota. Wisconsin had been working on some, Iowa and also Utah.
I would suggest that possibly you bring some of those people in as part of your process, looking at what they have done, what challenges they had with developing their legislation and maybe learn from their experiences.
Mayo Foundation is both a provider or a generator of electronic data interchange transactions. As part of our initiatives, as part of the Mayo Foundation, we are also a large payer.
So, we are also at the other end of the spectrum in being able to also receive transactions back into our system.
We have found that the X12 transactions do meet all of our data needs in a very good fashion.
Some of the concerns that we do have with some of the other standards, we do have concerns with the national provider ID number.
We had some concerns with the two digit location suffix. I was happy to hear today that they are considering separating that issue.
As a provider with thousands and thousands of providers, having that suffix be different from provider to provider to provider for the same location gets to be a very, very difficult task, a logistics nightmare.
That is why I was very happy to hear today that we were looking at separating those two issues. Then I had some concerns is, is two digits enough to identify the location for the purpose that this identifier is going to be used. So, we do look to have that addressed in the future.
Uniformity for us is extremely important, not only in the standards but in the data dictionaries. There are efforts in progress within the various standard setting organizations to develop a common data dictionary that will address some of the consistency issues, between the standards, for example, with the HL-7 standards, the X12 standards, NCPDP standards and so forth. That is also a very good effort that is going on right now.
We also would like to make sure that you pay attention to the schedule for implementation. It is important to keep things moving along, but also realize that there is significant effort for some organizations to make that switch, and look for some guidance on what does it mean when we want to do these types of transactions that are mandated and how are the penalties and so forth going to be enforced in the future.
These standards should be developed and maintained within an organization like the American National Institute, that has developed standard development processes. X12 adheres to these.
Through this committee, all stakeholders' business needs and requirements can be met in an open environment. All standards, new and changed, require consensus approval for the full ASC X12 membership. So, it is a very open and public process. Thank you.
DR. LUMPKIN: Thank you. Questions?
DR. IEZZONI: This is a fascinating panel. Beginning at that end of the table and going to this end of the table, we have some disagreements, maybe.
There was just one that I wanted to see if the two of you could talk about, Mr. Arges and Mr. Beatty.
Mr. Beatty, you have obvious enthusiasm for the ANSI 837 forms and are just really excited about them and implemented them with increasing efficiency as you moved along.
Mr. Arges, I thought I heard you make a plea for not having that be the sole format that this would go into.
Before I have the two of you talk on that, just glancing at the Lewin report that Mr. Rode was talking about, the first page that I turned to talked about costs, and the cost of implementing and the fact that this will be very costly for some organizations, that I am sure that Mr. Arges is concerned about.
How do you factor all of this in, from an organization like Mayo to the hospitals that you represent and the cost issues?
MR. ARGES: Can I go first? I think ultimately the end game is that we do want to be in an X12 environment. Our concern is that presently many of our members are very comfortable, particularly with the billing transaction, with the flat file that they have been accustomed to. They have had a very high success rate.
What they are concerned about is not so much the success of the transmission -- I know Gary has talked about what Mayo has done.
It is more along the lines of the data content, and that the data content be properly used and be properly applied as part of the process.
If I have to customize and mat each of my requirements, again, to each payer, then I have done nothing and my administrative costs continue to go up.
Until we have a clear idea about what components of the data elements need to be a part of the process, then you can start building around and moving people toward some of the X12 transactions.
Again, I think there should be a series of criteria that need to be established as well, to make certain that people understand what it is that they are implementing, making certain that what has been designed is being applied in the proper manner.
Implementation guides are being developed within X12. Until those are complete, until the industry has had review, I just caution the industry to move directly into that.
I would agree with Dan, that we probably need to start with four core transactions as part of that process.
MR. BEATTY: When we looked at implementation -- what we did was brought in a software product. We brought in a PC level product to do our translation to and from standards. We spent $5,000 to do that.
We were up and running without any internal work in three days.
MR. ARGES: Let me caution you on that statement.
MR. BEATTY: I will qualify those statements.
At the same time, there are other options that people do select, depending on their business practices, that are more challenging with patients.
Whether you choose to be on a main frame -- a lot of organizations still feel they need to be on a main frame for a lot of good reasons. Those solutions do cost significantly more.
Typically I always use the six figure range, about $100,000, to bring in the software and so forth, the training, to implement those things.
At the same time, I think as the information technology environment goes more and more to client server environment, that we have found that the client server environment is actually the better environment to do that.
We have found some solutions that are available over the internet, where providers can go out onto the internet and actually down load applications off the internet, that allow them to enter their administrative transactions and file them through a clearinghouse to the payer and back and forth.
That doesn't cost them anything other than per transaction, without having to make any investment in learning the standards and the technology issue behind that; essentially putting the whole process into a black box and keeping it away from the provider.
That is the way it should be, and that is some of the services that clearinghouses will provide, as we move forward in the selection of a process.
If providers can submit the information electronically and do so, they can file it to a clearinghouse that will handle the translation for them, again hiding all the technology from the provider community and the costs involved.
DR. LUMPKIN: Let me perhaps get a little bit of clarification on what you just said. Now you mentioned in Minnesota, what are the standards that are now required?
MR. BEATTY: Minnesota requires the use of the X12 standards, and is very closely in line with the regional recommendations from the work group for EDI, and their white papers.
DR. LUMPKIN: So, in your operation, you essentially took whatever you are doing and plugged it in through a PC to get out the X12 standards.
MR. BEATTY: Correct.
DR. LUMPKIN: So, you really have not changed your internal operations at all.
MR. BEATTY: And we didn't need to. A lot of providers don't need to.
DR. LUMPKIN: You have reached standardization without simplification.
MR. RODE: If I can make a comment, on EDI, if you can standardize the data, the EDI software does the translation from the applications within one entity across the EDI translation, across the network to a receiver, who also translates it back into their own application.
It is kind of like the way the United Nations does translation during simultaneous activity. There is somebody up above translating from Swahili to French, from French down to German.
What the two people on each end hear are German and Swahili. They don't hear any of the French translation. That is essentially part of the way that the EDI process works.
It allows you to send the information through without having to change the applications, as long as you have the data correctly identified and have the data to send, or have a place to receive it if you need to receive it.
DR. LUMPKIN: My question is, and this may be paraphrasing somebody or other who said it, but I think my goal is to standardize in order to simplify. It is not just to standardize.
From what I just heard you say, Gary, I didn't hear that what you did had any impact upon the clinical practice in your institution.
MR. BEATTY: It did in that we don't have to chase down the claims. Again, we are building an electronic commerce environment that goes beyond claims. Claims is just one piece of this whole puzzle, and putting it together.
The claims helps get the transactions out of the door. We don't have to deal with the paper and sorting and mailing and postage and all those kinds of issues.
But the real benefit to us is the process when we receive information back in, like the electronic EOB, where we can post that information electronically back into our system, and automate that process, and reduce our administrative burden for re-entering that information.
There are other transactions like the claim status transaction that we are implementing with HCFA, that will streamline the process of handling any question regarding the adjudication process that the payers may have, again eliminating the need for us to play telephone tag or play with snail mail back and forth between the payer and provider.
Those are the things that we are looking at as moving information more efficiently and living in a much more expeditious manner.
Mr. ARGES: I was just going to comment, too, Gary mentioned some costs. I would also caution, though, that to be EDI ready requires some heavy investment.
If I am not mistaken, it was my understanding Mayo may have spent $250,000 just to become EDI ready. That means telecommunications packages and other components that also need to go part and parcel with the process.
Again, I think it is important that we stress the content functions as opposed to the success of the communication. Being able to receive a communication or send a communication, while you may have success in that regard, you need to be able to understand what it is that you received and processed, as part of that niche where we need greater standards.
DR. LUMPKIN: Did you have a follow up?
DR. IEZZONI: Yes, but it is on a slightly different topic.
DR. LUMPKIN: Does anybody else have anything on this topic?
DR. SCHULTZ: Can I make a comment? Part of your earlier question was wondering how you are going to be successful or knowing that you are successful after it is all over.
That is a really tough question because there are a lot of simultaneous interventions going on at the same time. So, you have a very complicated experiment here.
I know we have asked the same question within our institutions at UHC and HL-7 asked the same question. So, we tracked costs of interfaces.
I know when we began in the period of 1985 to 1990 -- and I was a CIO at the time at the hospital at the University of Pennsylvania -- interfaces would regularly cost $50,000, $100,000 for just connecting a new system.
We think those costs have now gone down an order of magnitude, so that they tend to be more in the $10,000, $15,000, $5,000 range.
We are hopeful, with the next generation of object technology, that they will be down at the level that George is talking about.
We do track interface costs for everything we implement within the consortium, so that we can see how much they cost.
There are sort of distal and proximal outcomes. A proximal one would be this cost. The distal one is whether or not it is plugged in play.
If it is plugged in play, then at least, whatever you have done, it means you haven't added to the cost of the institution.
As to whether or not you have alleviated time that would have been spent on administrivia and been able to shift it toward care, that is going to be really hard to track.
We try to track that, but we look at it in more global terms under what we call an operations improvement bench marking program, where we will take a whole bunch of these things, look at them, and then follow them, track them over two or three years.
It seems to me that one of the things you are going to want to do is to look at some institutions up close that have the ability to track those kinds of things, both large institutions and small institutions, which is I think what the issue is here.
We have some new small institutions that don't have the wherewithal that some of our large institutions have, to even be able to see these changes.
Then to look at them, one, whether the proximal costs have been diminished down to zero as you implement these transactions, and two, whether in fact they have achieved nearly plugged in play.
Three, I think you will try to estimate whether or not you have accomplished reductions in administrative costs. That, I think, will be a very tough challenge.
DR. IEZZONI: While we have Mr. Arges here, I wonder if he could weigh in on the coding issue. I thought I heard you say in your comments that it is a problem for some of your members that there is disparate procedure coding. Do you have a preference?
MR. ARGES: I don't know necessarily that I have a preference. I would like to have one to be used by the institutional provider community as a matter of routine data collection.
It is when I have different coding nomenclature that I have a problem. I think we need to take a look at which one provides us with the better information, that it is applicable to the institutional setting, that it is updated frequently, that there is a form to discuss the issues if need be, and go from there.
DR. LUMPKIN: Can you clarify, when you say institution, you mean inpatient and outpatient?
MR. ARGES: Inpatient and outpatient.
DR. BRAITHWAITE: I just wanted to ask Sam, who said that 95 percent of your members claim to use X12 and HL-7, how that can be sort of reconciled against what we heard this morning, that only about 20 percent of organizations are using EDI.
From the figures that HCFA has published before, that outside of HCFA, the amount of use of X12 is in the under-10 percent range.
DR. SCHULTZ: First of all, I wasn't here this morning, so I don't know where that data was accomplished. One of the things we do is to regularly survey our CIOs who are implementing specific systems and specific transactions to look at, one, whether they use a proprietary or another standard, which one they do, which company they used, how they did it, what it cost and so on.
We try to track both direct and indirect costs across all of our institutions. I will be glad to share the data with you.
I don't know where the other data came about and maybe there is a big difference between our academic health centers and other organizations.
I suspect what I am hearing between the two Georges -- George and Gary, excuse me.
DR. LUMPKIN: How many members do you have?
DR. SCHULTZ: Institutions?
DR. LUMPKIN: Yes.
DR. SCHULTZ: It depends on how you count, but in terms of universities, 150 or 170, depending on how you count.
DR. LUMPKIN: That is 150 out of the 5,000. You don't include the hospital in Sandwich, Illinois.
DR. SCHULTZ: We are beginning to, you see, because our members are acquiring hospitals. That is why I really hesitated to give you an answer. Now that Cook County and places like that are in there, we are changing some of our --
DR. LUMPKIN: That is a 10-bed hospital, and I am sure that they haven't gone to that. I don't think they have a PC there.
DR. SCHULTZ: That is really part of the issue, is what are you trying to hit at this point. What is it that you are trying to harvest, and we have the same kind of issue, not letting perfect get in the way of good, trying to move the whole agenda as best we can.
Anyway, I would be glad to share the data with you, and we do have it. I suspect it is in that large institution/small institution gap that we are seeing the difference.
DR. COHN: First of all, I actually want to thank Gary for his suggestion about recognizing that there are a number of states that have already been taking some leadership in this.
I do want to recommend to us that we should try to get some of those states in to talk about their experiences. That was a very good idea. I would like to know a little more.
George, I actually had a question for you. I know you are here in sort of an awkward role. On one level, you are the chair of the NUBC and obviously very much vested in the UB-92, especially the ideal data set as created by it.
You are also here representing the many affiliated hospitals that are members of the American Hospital Association.
Now, taking off your hat as the chair of the NUBC and only representing your hospitals, how satisfied are they really with the UB-92, as they are having to use it, recognizing that most of them are dealing with multiple different subsets of that standards. I thought you said very freely. From your view, are they very satisfied with what is currently going on?
MR. ARGES: I think they are very satisfied with what is currently going on. Again, the design of the UB-92, again, it was designed with some level of flexibility in terms of adapting to different reporting requirements.
What they are concerned about are really people who operate outside the NUB process, who compromise the data set, or who aren't willing to use the data set in its entirety.
Then they have to go through special coding, special reporting. That is when they have problems.
DR. COHN: What percentage of bills are sent out using the data set as has been suggested by the NUBC?
MR. ARGES: I don't know if we have a statistical summary. If I look at, again, what has been done with the Medicare program, Medicare basically comes to the committee, they present their case for data requests.
Again, by and large, we have had a very high success rate, close to 97 percent of claims moving electronically via the Medicare program.
Medicare represents about 40 percent of our members, in terms of market share. So, it is a very high percent.
I think a lot of the Blue Cross plans, as well, utilize much of the data set in the same fashion.
I know many of the members on the committee who represent payers support the UB as a whole. Again, we haven't done it without any teeth behind us. We have done it on a voluntary basis.
But again, it is trying to correct problems when they do occur. It is an ongoing process. Again, it is open.
MR. BEATTY: I would just like to add to that. Both the national standard format and the UB-92 format, as they are defined, do a very good job.
It is a problem of the people going outside the box and changing those standards to meet their needs. The challenge is that most people do that.
We have, unique in our environment, we have unique implementations for just about every single trading partner that we do electronic data interchange with, except for those we are using the X12 standards, because they are more rigid in their structure and definition.
I am not saying that is always the case, but it is less frequent.
DR. LUMPKIN: Would you define that as meeting their needs or their wants?
MR. SCANLON: In terms of the -- again, this is a question about timing and the overall process. You recall that the HIPAA asked the Secretary to adopt standards in these nine categories and supporting standards within 18 months, and then allows an additional 24 month time period for implementation for most participation and then another 12 months for small plans.
Now, virtually every one of you seemed to conclude that in most of these areas outlined in the law there were some existing standards or some candidate standards that obviously should be looked at first.
You also mentioned that a fair amount of migration would be needed to get to those standards. What is your sense of this time period of migration, from the point when the standards were adopted and then there is another minimum of two years, it looks like, to the point of action of adopting them.
Is that the period of migration you are thinking of, or is it longer than that, or shorter than that?
MR. BEATTY: That is a real tough question to answer because it varies by the organization, their management structure, the systems that they are utilizing, whether they are home grown systems, whether they are purchased systems.
We heard earlier that one person is using a system that none of their programmers are on this planet. It is a real subjective type of thing as far as implementation time.
I know when I was with a payer and the first time we implemented electronic data exchange, it took us about a full year to implement our first transaction. So, it did take a considerable amount of time for the first transaction.
The second, the third, you know, as I mentioned earlier, took a considerably less amount of time. It moves right along.
Again, from the provider's perspective, if you don't want to utilize that, and you want to utilize the clearinghouses and so forth to make that happen, it can happen very fast.
The clearinghouses and the value added networks are ready and willing to take on those challenges.
DR. SCHULTZ: I think the issue is the steepness of the curve. That is, the first couple are very costly and I think if there were carrots to get more, especially less -- to get hospitals that had less resources under that curve quicker so that they were further down along the line in transaction, it would speed things up.
That is really, I think, the issue that we see. Once we see a hospital adopting one transaction and then a second and then a third, then all of a sudden they are quick adopters of new transactions.
So, I think managing that front end of the adoption curve in some reward fashion is what needs to be designed, so that it appears to be a win for that institution rather than a net loss over some period of time.
MR. RODE: I was going to say, I think two things. We suggested a longer period, to go not to 24 months but to move to 30 months or 36 months, partly to accommodate that.
I also think it is important that if the public -- in this case the health care industry -- knows where you are headed, if we know that you are going to go to the X12 standards, or the use of the X12 standards as your base, we can begin the education process.
Most of these institutions have no education at all. They have been waiting to see, what way is it going to go or is it going to go at all. They are waiting for that.
If we can begin some of that education, some institutions are using EDI for none of the transactions we are talking about.
They are using them for purchasing and accounts payable and a variety of other things, who can take that technology and move quickly.
Some are not using anything at all. Our organizations, maybe with the exception of Gary -- part of our responsibility is education.
If we know where we can begin to educate our members, we can do a lot to help that. But if you don't let us know and it is down to, well, it is two more months and we are still not sure exactly how we are going to go, there is going to be a long implementation.
So, there is a lot that can be done if we can stage the information in such a way, and build on it to allow us to do the training that helps.
The incentives are something we have also suggested, but we have made marvelous improvements in physician billing in fairly short periods of time.
I tend to think that if we can do this as a group, we can meet the need. But I do think we need more than 24 months, to be honest with you.
MR. SCANLON: What do you have in mind in terms of incentives. Are you talking about tax incentives?
DR. SCHULTZ: That is the kind of thing I was thinking of, since we seem to be driven by those kinds of things these days and we are in an environment where those seem to be part of the purchase in new process and business change.
It seems to me that is one way that would provide appropriate rewards for a business that is undergoing substantial change, and maybe rewarding some other consequent changes, the business process change, along the way.
MR. ARGES: Can I comment a little bit on your question as well? I think it goes back to the whole issue of defining, again, the data content associated with each transaction.
It might behoove us to take a look at the four core transactions to define each of the data content components that make that up, identify where the origins of some of this data basically comes from, and look at the role of each participant to the data collection process.
Some of the information providers report on the claim information that could come from the enrollment or eligibility files, some from remittance files.
The idea is to make the system simpler. The idea is to be able to share data bases from one organization to the other organization and exchange that rather easily, and build from that a more complete picture of what is actually taking place, but not necessarily to recreate the data gathering process within each setting.
DR. STARFIELD: I have a question for you, George. Early in your presentation you made a distinction between administrative financial data and clinical data, and not just to focus primarily on administrative financial data.
You know, the more things change, the less distinction I think there is between administrative data and clinical data.
Gary actually gave some examples of how administrative data are being fed back into clinical decisions. HETUS is an administrative data system and that has real clinical application.
What is the distinction you make between it? I can see the billing versus clinical. I am having a little trouble seeing the distinction between administrative and clinical.
MR. ARGES: I think these are separate transactions that have been designed, really, to take into account some level of accountability within health care in terms of the financing of the health care system.
These are really summary documents that represent key elements for information exchange.
Clinical information systems are designed moreso, I think, for the practitioner, being able to provide them with information about the treatment and care of the patient.
A lot of those elements need to be abstracted and pulled onto some of these other information systems. One shouldn't -- we shouldn't have to wait until we get a complete clinical information system in order to work on these other ones.
These other ones are so much further along, we could easily put those into place and start working on some of the more operational issues associated with clinical information systems, a building of a more complete medical record, being able to share log information with the lab and being able to exchange that from one department to the other.
Again, these are for external reporting, moreso than between practitioners.
DR. STARFIELD: I don't know, maybe some of the others of you have reporting. There is a lot more external reporting now that involves clinical data. HETUS is a good example.
DR. SCHULTZ: I think you are absolutely correct in terms of how you utilize the data. The lines are grayer and grayer.
I think the concern is that standards are set for data and then transactions are made easier, without worrying about whether it is clinical or administrative.
In fact, there are two standard development organizations attempting to standardize the same thing, that there be a process where they talk and they agree to have a convergence standard, so that it can be both and serve both needs.
That appears to be working, finally, now that we have a basis for standards groups working together now that we didn't have two years ago. But you are right.
MR. BEATTY: I just want to touch on a couple of items. What are some of the efforts that are going out, for example, from the Centers for Disease Control, and the building of the national data base for the immunization of children and so forth, and the passing of clinical information and counter type of information in that way, to build a data base that can help improve the quality of care with the United States. Those are very good endeavors.
The second thing is, what Sam mentioned, the Health Care Standards Board, the American National Standards Institute Board, is working on resolving issues of overlap where they do occur. That process is working very well.
As a matter of fact, NCPDP and X12 are currently working through some issues of overlap and I am very pleased to be part of that process. I think it is working very well.
I think HSBE is playing a very good role in that endeavor and I congratulate them.
MR. RODE: I think some of the clinical standards are merging into the administrative. We are seeing it in different places in the country. If you go across the country, it is so different, from Southern California to New England.
HETUS could easily find its way into one of the transaction sets that we see today, some of the coding systems may find their way out of the transactions that we have today.
At the point we are at right now, we are trying to go through an administrative simplification process, partly to lower the cost of what it is taking to do these nine transactions, and begin to set up a model to do new transactions.
I think the concern that George is addressing and certainly we are, is if we are waiting until we get everything, we would still be waiting, because this industry just moves so fast.
If we waited to make sure we were absolutely sure the internet was the answer, we would have a long wait.
I think it is important that we build the standards to begin to handle those. A few years ago we couldn't handle physician ID numbers and we couldn't handle some of the data that HCFA wanted on several occasions.
Once we were even able to convince HCFA that we wouldn't carry the data and they shouldn't carry the data. But more and more, those lines are changing, especially with managed care.
I think work on the X12 and others are trying to make sure that, when they are accommodated within the standard, we try to hit this uniformity requirement, and the people using the standard know how to use it.
DR. LUMPKIN: Your organization includes financial folks.
MR. RODE: Right.
DR. LUMPKIN: From more than just the hospital arena. Could you perhaps share with us a little bit of your insight on the impacts of administrative simplification on the home health and the nursing home arena. Are they ready?
MR. RODE: The home health area is kind of like the nurse who was up here previously. These are small shops, using in most cases PC-based software, if they are not just putting information on a sheet of paper and sending it out to an out-sourced business office.
Some of the comments before about some of the data sets, why in the home health industry we have two data sets that apply.
Some home health agencies are owned by hospitals that would like to have a system that could handle both of them in the same billing arena, but can't because the way you do the transaction is even different beyond just the data that is transferred.
In the skilled nursing facility there is going to be a difference, whether they are in a chain or they are a stand-alone situation.
So, it is similar to George's comments about hospitals and Sam's about university facilities, because we are in an industry that is very quickly, in some parts of the country, integrating to an extent.
Yet, we have to keep our financial transaction systems separate because we don't have the commonality of data.
We don't have the ability to have one data base and say, okay, this is a home health bill, send it out, this is a long term care bill, send it out.
We keep our systems separate because of the differences you have already heard.
For our small members, EDI gives the promise of the kind of solution that Gary talked about, that can be done for $4,000 or $5,000.
For our larger members it is a question of, well, do I scrap what I am doing now and go to something new, or do I wait a while and see whether a new technology is coming in.
Again, the technology is partly going to follow what you do. If you decide that you are going to accommodate large facilities and the small ones are going to have to go somewhere else, all of a sudden there are going to be clearinghouses that we don't even know exist that are going to come on the carpet.
If we go to large models, there are going to be people with new EDI translators that we haven't even seen before. The industry is moving that quickly.
They are watching you. What you do is going to generate where the technology goes. Remember, we are a fantastic market right now.
We have got well over two million entities that are going to be affected by this. So, what you do is also going to drive some of the technology that goes forth from this point.
DR. BRAITHWAITE: Especially George, you were mentioning the concept of saying, if we standardize a claim, then all payers have to accept the claim. There is this sort of goal that will prevent people from implementing local codes and special deals and all this sort of stuff.
In fact, isn't it true that if we implement a standard claim, that there will be special deals all over the place anyway.
Those pressures aren't going to go away. Somebody is going to figure out how to do it through the back door or you have to send in a special piece of paper along with your electronic claim or some other transaction?
MR. ARGES: Not if you detach, I think, the requirement be attached to providing the additional piece of information.
I think if you are basically saying you provide the information required to fulfill the transaction, then once you have fulfilled the transaction, then you should be obligated to receive payment as part of that process.
DR. BRAITHWAITE: So, you are taking away the leverage of the payment.
MR. ARGES: Exactly, and if I wanted to provide additional pieces of information based on the contract, then that is up to me to assume as part of the requirement.
But I wouldn't necessarily be coerced into providing the additional pieces of information just to receive payment.
DR. BRAITHWAITE: I think that is a very important distinction. Does anyone else have a comment on that?
MS. COLTIN: I have a question following up on the vision conversation that we had earlier. I know that there is clearly a trend for managed care organizations to begin to pay hospitals and other facilities under capitation arrangements, rather than fee-for-service type arrangements.
I am curious whether, within your memberships or your own organizations, you have seen a tendency to want to request for the managed care organizations, to want to request from their departments clinical data sets rather than transaction data sets, as a way to look at the services that are being delivered under that capitation arrangement.
For instance, capitating emergency rooms, but then wanting to get the emergency room data sets as opposed to a bill, so that they can look at quality of care issues.
Likewise, the transaction can fall out of that data set, but the opposite is not necessarily true.
Are you seeing any of that sort of thing happening? I know that it has happened in the laboratory arena.
There are some large managed care organizations that, within their arrangements with laboratories, are getting clinical data back from the laboratories, as well as billing information. Are you seeing this in your facilities?
MR. RODE: I would have to answer that, again, it varies across the country. Contracts between the managed care entity and the hospital or the physician's office do demand certain types of information be sent.
Some of the standard transactions allow for some of that data. There is the ability to do capitation now. In the X12 standards is the ability to do authorizations and certifications and a variety of uploading and down loading, who is covered, who isn't, and who should be the gatekeeper, if there is that sort of a process.
Certainly, if we had data sets that could include all that other additional information, that would be great. As has been pointed out several times, we just don't have those yet.
So, what the institutions are doing is trying to get as much information in electronic format as they can, based on what we have today.
In the cases where we don't have that information, supplying it on a paper basis or having, in the best possible world, come in and look at it in our facility so that we don't have to send anything.
As the gentleman earlier this afternoon said, an indicator saying that the data is on file.
MR. ARGES: I think there are a lot of situations like that, that are cropping up all over the country. I think it is increasing.
I think it is done more on a limited study, where they may want to look at a particular practice for a given period of time.
So, these aren't necessarily routine transactions that are taking place on a continuous basis, but a short ad hoc study that is taking place, or a study that is done on a periodic basis rather than on a routine basis.
I do think that the dynamics of managed care, the incentives involved with it, do change the frequency of reporting the information within the data set.
I think the data components that are core to defining, let's say, an encounter, are similar.
DR. SCHULTZ: We see a lot of interesting things going on. First of all, to stay in the marketplace, in many of our marketplaces, we stage the marketplaces for our members so that they are aware of what marketplace they are in and what they can do about it.
On the west coast, for example, they have to go into a lot of capitated arrangements. They will do it without knowing whether they are making money or not.
That may sound strange to a lot of people but, in fact, if you want to exist as the university of whatever medical center, you have to do that.
So, we end up doing funny things, like deciding, as we become more and more capitated, more and more managed care, will we continue to use billing systems as de facto clinical systems, and for how long do we wait for the valhalla of the computerized patient record.
So, that is exactly what we are doing. We are using UB-92 detailed charges as proxies for care, pharmaceutical charges inducing through the pharmaceutical standards and the other standards the details of the care process.
We expect that to go on for another administration, at least. So, in order to compete and survive in a managed care marketplace, we expect both those standards that you have already issued, and those that you will continue to issue, as being the basis for de facto data formatting that we use in estimating whether or not we are providing the right care, and at what cost, and what the quality of care is.
I think that is the most important issue for us, is assuring that, even though your focus now is on administrative issues and financial issues, that this harvest of cost is going to reach an assintote, and the issue of the reform of the delivery of care as it is delivered is the next wave.
We have to be prepared to have the data system to collect that kind of data. That is really where the prize is, we think.
DR. LUMPKIN: Thank you very much. At this point, we are going to adjourn the work group until tomorrow and we are going to go back to the subcommittee.
DR. STARFIELD: Thanks very much. What we are going to do now is basically adjourn the meeting. We don't have a quorum for subcommittee business. So, we will adjourn.
However, I will ask the remaining members to stay so we can just discuss some logistical issues. Thank you very much.
(Whereupon, at 5:10 p.m., the meeting was adjourned.)