Department of Health and Human Services
NATIONAL COMMITTEE ON VITAL AND HEALTH STATISTICS
Subcommittee on Standards and Security
December 10-11, 2002
Washington, D.C.
_______________________________________________________________________________
- Minutes -
The Subcommittee on Standards and Security of the National Committee on Vital and Health Statistics (NCVHS) held hearings on December 10-11, 2002, at the Hubert H. Humphrey Building in Washington, D.C. The meeting was open to the public. Present:
Subcommittee members
Others
· Don May, Amer. Hosp. Assn.
EXECUTIVE SUMMARY
NCVHS is the main public advisory committee to the U.S. Department of Health and Human Services (HHS) on national health information policy. During the morning of the first day, the Subcommittee on Standards and Security focused on the HIPAA electronic transaction standards final rule. Presentations included the designated standards maintenance organizations’ (DSMO) annual report and recommendations, the Centers for Medicare and Medicaid Service’s (CMS) initial analysis of the Administrative Simplification Compliance Act (ASCA) database, and perspectives on improving and stabilizing the HIPAA process. In the afternoon, members discussed the scope of work for the cost/benefit analysis for migration to the Clinical Modification of the International Statistical Classification of Diseases and Health Related Problems, Tenth Revision (ICD-10-CM) and ICD-10-PCS, Procedure Coding System, the latter for in-patient hospital procedure reporting. Members also discussed the January hearings on provider and payer needs for code sets for complementary and alternative medicine. The next day the Subcommittee considered the scope and criteria of their work on patient medical record information (PMRI) terminologies.
Ms. Ward presented the annual DSMO Progress Report, addressing three major areas of work that the Steering Committee undertook: DSMO change request volumes and status for the implementation guides adopted under HIPAA, non-change request initiatives undertaken by the DSMO’s, and specific DSMO requests that warranted additional consideration due to their nature. She noted the DSMOs were also involved in working with codes that maintainers sought to have named as standards, supporting HHS in the NPRM comment period, NCPDP and NDC changes, and discussing new requests through the DSMO process that potentially might affect all covered entities exchanging transactions.
The Administrative Simplification Compliance Act Database and Best Practices - Subcommittee Discussion
Ms. Trudel reported that some 540,000 requests for a one-year delay in the implementation of the HIPAA transactions and code sets were submitted. Most people either indicated implementing HIPAA would cost them less than $10,000 or between $10,000 and $100,000, or they reported they didn’t know. About 158,000 said they were using or would use a contractor or vendor; 215,000 were undecided. Asked the reason for delay, 250,000 needed more information about the standards, 254,000 waited for a vendor, 368,000 needed more time, 226,000 had to test, 185,000 waited for addenda and needed additional clarification on the standards. Members noted issues that needed to be researched. Ms. Trudel reported that the final rule was near the end of the review process and on their post publication schedule for the end of December. Dr. Cohn said if it wasn’t out by January, the Executive Committees needed to send a letter to the Secretary, but he hoped that wouldn’t be necessary.
Improving the HIPAA Process
· Jack Rovner; J.D.; Michael, Best & Friedrich, LLC
Mr. Rovner presented the Secretary’s Advisory Committee on Regulatory Reform’s work on the HIPAA transaction rules. CMS adopted a defined schedule, with dates certain and a predictable, annual cycle for the issuance of updates of code sets, data elements and standards. Other recommendations concerned rules that lacked four essential elements (sender, receiver, content, and purpose requirements) that usually appeared in a standard, causing confusion, uncertainty, and controversy. The advisory committee also called for clarifying the regulations with respect to direct data entry exceptions. Mr. Rovner reiterated that the rules were needed as quickly as possible along with a predictable schedule and date certain, so everyone knew every year if and when a rule change took place.
Improving the HIPAA Process
Ms. Ward noted that a WEDI work group’s primary issue was a predictable schedule for implementing these transactions. A triage system, where the approval process was consistent with the magnitude of the change, was suggested for processing the changes and the transactions evolution. The group recommended not having more than two versions of the standards in production simultaneously. Noting there’d be some parallel between versions, the group cited the need to define structure and dates. Ms. Ward echoed Mr. Rovner’s comments, and stressed that the time factor was problematic and a big concern.
Scope of Work for Cost/Benefit Analysis for Migration to ICD-10-CM and ICD-10-PCS - Subcommittee Discussion
Ms. Pickett reviewed and amended the draft of the cost/benefit analysis scope of work that captured comments made at hearings earlier in the year and a discussion at the last full Committee meeting about issues and concerns that should be part of the scope of work. Changes will be e-mailed to members; comments are due December 18. A contractor will be identified who will discuss their plan with them at the January meeting. The end date for the report is a week before the May meeting. The Subcommittee will then review the report and decide whether to make a recommendation to the full Committee on an NPRM or have additional testimony, thereby delaying a recommendation until the September meeting.
Preparation for Complementary and Alternative Medicine Hearings in January 2002
· Richard Nahin, Senior Advisor for Scientific Coordination and Outreach, National Center for Complementary and Alternative Medicine
Dr. Nahin noted that the National Center for Complementary and Alternative Medicine primarily funded clinical research, reflecting the fact that alternative medicine was available to and used by the public. Alternative medicine was divided into five domains: mind/body medicine, traditional systems of medicine, biologics, energy-based systems, manipulation and other body-based systems. The Center used health services research and other epidemiological studies to set their research agenda. Some 150 clinical trials were in progress. Dr. Nahin noted that practitioners of a traditional system of medicine diagnosed in different ways than the conventional physician and had different diagnostic categories. He added that one thing the Center looked for was correlation in diagnostic schemes for certain diseases in which conventional diagnosis systems weren’t accurate and that were a diagnosis of the absence of anything else. He said an alternative medicine diagnosis overlay might enable more homogeneous groupings of patients.
DAY TWO Report on the Consolidated Health Informatics Initiative, Steven Steindel, CDC Dr. Steindel reported on a Consolidated Health Informatics (CHI) Initiative work group that was charged with prioritizing terminology domains important to electronic health records based on how quickly work groups could be established and recommend terminology. CHI’s goal was the interchange of healthcare data among federal agencies. Four priority domains were developed: domains with reasonably apparent terminologies, so work groups could select appropriate terminologies to recommend; domains very important to the electronic medical record with terminologies less mature; domains where terminologies needed extensive work or didn’t exist; and terminologies essential for the electronic health record that were maintained by other groups and should be adopted as federal interchange standards (e.g., HIPAA). Members agreed that CHI and NCVHS could share a synergistic relationship. Discussion on Scope and Criteria for PMRI Terminologies - Committee Discussion
Dr. Sujansky reviewed an update of the second draft of the document, seeking agreement on a framework enabling the Subcommittee to consider the criteria for selection and fold them into a questionnaire as the first part of their information gathering activities. The Subcommittee agreed that the intent in characterizing Groups 2 and 3 wasn’t to select terminologies that met the objectives of standardized terminologies for PMRI. Members concurred that the core group would meet the objectives of interoperability and comparability that were the ultimate goals. The other groups acknowledged other important terminologies that had to be accommodated and that provisions were needed so those using these other groups could adopt core terminologies for clinically specific data capture and analysis. Members sought consensus on whether each criterion was useful, and if it was quintessential or desirable. This input and discussion of the criteria will be reformulated as candidate questions for the survey and circulated to members the following week.
Review Revised Scope of Work Draft for cost/benefit analysis for migration to ICD-10-CM and ICD-10-PCS - Committee Discussion
Members completed a final reading and unanimously accepted the draft scope of work for the cost/benefit study on the impact on the healthcare industry of moving to ICD-10-CM and ICD-10-PCS.
DAY ONE
Dr. Cohn welcomed everyone, especially Brady Augustine who would leave the Committee after that meeting for a new position at CMS, to the first of two days of hearings of the Subcommittee on Standards and Security of NCVHS. Noting this is the main public advisory committee to HHS on health and national health information policy, Dr. Cohn explained that the morning’s focus was on the HIPAA electronic transaction standards final rule. The DSMO would present their annual report and recommendations. There would also be an initial analysis of the ASCA database CMS was developing. The Committee’s responsibility was to identify implementation issues derived from that database and publish reports on effective solutions to compliance problems. Afterwards, the Subcommittee would discuss improving and stabilizing the HIPAA process, deliberations that will continue at the January hearings. That afternoon members would define the scope of work for the cost/benefit analysis for migration to ICD-10-CM and for ICD-10-PCS, the latter for in-patient hospital procedure reporting. Members also would discuss the January hearings on provider and payer needs for code sets for complementary and alternative medicine. The next day the Subcommittee would consider the scope and criteria of their work on PMRI terminologies.
Ms. Ward presented the annual DSMO Progress Report as required by the Memorandum of Understanding (MOU) signed by all the DSMOs and HHS. She addressed three major areas of work that the Steering Committee undertook this year: (1) DSMO change request volumes and status from July 2001 to November 2002, (2) non-change request initiatives undertaken by the DSMO’s, and (3) specific DSMO requests that warranted additional consideration due to their nature (e.g., standardizing insurance cards).
Ms. Ward noted that 2002 was an extremely busy year for covered entities in the process of implementing X-12N and NCPDP transactions adopted by HHS for HIPAA. With the increased awareness and activity, the DSMOs had been busy trying to accommodate reasonable requests for changes to the implementation guides adopted under HIPAA. They were also involved in working with codes that maintainers sought to have named as standards, supporting HHS in the NPRM comment period, NCPDP and NDC changes, and discussing new requests through the DSMO process that potentially might affect all covered entities exchanging transactions. Ms. Ward said the DSMOs believed they’d been successful and that their recommendations would continue to serve the healthcare community and advance HIPAA implementation.
She reported on DSMO change request volumes and status from July 2001 through November 2002. Ms. Ward noted the July batch was the first monthly batch of requests that weren’t part of the fast-track process. The July 2001 through April 2002 batches completed the process. A total of 143 change requests were entered in that time period. Washington Publishing withdrew nine requests that were incomplete or duplicates. Submitters withdrew another 52 prior to the discussion period. And 82 completed the process. A submitter withdrew one appeal. Five were denied. An appendix to Ms. Ward’s report detailed the completed change requests, including DSMO responses.
Some 31 requests were classified as modifications necessary to permit compliance with the standard. HHS delineated necessary items as: (1) something in the adopted standard or implementation specification conflicted with the regulation, (2) a non-existent data element or code set was required, (3) removal of data content no longer supported by the healthcare industry, (4) lack of a data element or code set critical to the industry’s business process and, (5) a conflict either among standards or within the standard.
Four requests were classified as maintenance (items that didn’t impact implementation of the transactions) requiring no further DSMO action. Maintenance requests follow the standards development organization (SDO) process.
Another 47 items that went through the consensus-building process or had implementation guides that met the requested needs were classified without change. Ms. Ward noted that submitters could appeal the disposition. Some 88 change requests from May 2002 through November 2002 hadn’t yet completed the process.
Ms. Ward reported non-change request initiatives the DSMOs addressed. The Steering Committee helped the Department prepare for and respond to comments on proposed rules for addenda to the X-12N version 4010 and changes for NCPDP and NDC. She noted this was a significant undertaking for the DSMO committees that consulted with their members throughout the process. The Steering Committee met two-to-three hours weekly, sometimes three times a week, responding to public comments on the NPRMs. Coupled with individual committee meetings and preparation for each meeting, for some DSMOs this became nearly a full-time job.
Noting significant challenges faced in the comment response process, Ms. Ward said X-12N addressed items they considered to be issues with the process. Once they’d completed their analysis, the Steering Committee would address these issues. Ms. Ward noted the significant effort required to make the NPRM process and subsequent final rule successful. As requested by CMS in a letter to the DSMO’s following the comment response period, the Steering Committee will address any issues around this process they feel require attention.
She reported the DSMO’s also worked to develop guidelines for naming new code sets as HIPAA standards. First, DSMOs believed prospective terminology maintainers should meet the guiding principles expected of all current maintainers of the selected HIPAA standards (e.g., improving efficiency and effectiveness of the healthcare system by promoting cost reductions for and improvements in benefits from electronic healthcare transactions). DSMO’s believed every effort should be made to consider the use of existing code sets in meeting business needs before adopting additional sets. They considered these requests different enough from data element requests that the public process should be tailored to this specific need. DSMOs were working on the final draft of the code set guidance and coordinating with Washington Publishing to add it to the DSMO Web site.
Ms. Ward said the DSMO’s were in the final stages of developing a charter for the Steering Committee that will formally document the committee’s operating procedures including its organization, officers, voting procedures, and meeting guidelines. The Steering Committee had been working to develop a contract with Washington Publishing Company for the maintenance of the DSMO Web site. The site had operated for two years and the committee felt strongly that there should be a formal written agreement between Washington Publishing and the DSMOs. Ms. Ward said they’d discussed multiple possibilities, but reaching agreement was a challenge because the DSMOs weren’t a legal entity. Some DSMOs had engaged legal counsel and continued to work through the details.
Another significant work item involved requests affecting all HIPAA stakeholders. While the vast majority of change requests adjudicated addressed data element or segment changes, Ms. Ward said several were more significant and potentially affected all entities involved in HIPAA. She said the DSMOs invested a great deal of time discussing them, both at the Steering Committee level and within individual DSMO organizations. She said the DSMOs believed that such significant requests warranted a separate resolution process. The Steering Committee will develop a process to consider such requests.
NCPDP submitted two change requests regarding the use of their standard. The first, which would allow use of the NCPDP standard for billing professional pharmacy services, was denied by the DSMO with the recommendation to continue current business practices. The second, which allowed use of the NCPDP standard for billing supplies, was supported.
Several change requests from the same organization and submitter requested adoption of new transaction sets under HIPAA: (1) adoption of the healthcare benefit coordination verification request and response transaction (X-12-269), a transaction that goes from a primary payer to a secondary payer to indicate EOB information so secondary payers don’t request paper EOBs from the provider, (2) the 277 unsolicited implementation guide or standard as an acknowledgement and for mandating electronic funds transfer and the use of standard ID cards, and (3) a request to replace the NCPDP standard for claims with a specification developed by a sector of the pharmacy community and not yet adjudicated.
Noting it was an honor to lead this group of dedicated, hard-working volunteers, Ms. Ward thanked the Subcommittee for their continued support and guidance. Remarking on the difficulty of bringing together diverse views and interests within a single organization, Dr. Cohn acknowledged Ms. Ward’s leadership and thanked her and the SDOs for their efforts.
The Administrative Simplification Compliance Act Database and Best Practices - Subcommittee Discussion
Ms. Trudel explained that ASCA gave people the opportunity to submit a request for a one-year delay in the implementation of the HIPAA transactions and code sets. They had to do so by October 15, 2002 and provide specific information that conveyed an idea of what was happening in terms of national compliance. The ASCA statute tasked the Committee with looking at these results and coming up with possible assistance to address compliance challenges. Over 500,000 requests came in electronically. Another 50,000 were submitted on paper. Ms. Trudel reported that they’d begun to see what this data told them about providers’ compliance problems and solutions.
She noted the data was still rough and the reports were hard to read. A number of people submitted multiple requests (some didn’t realize they’d already received confirmation; others corrected mistakes and made changes), and the contractor had un-duplicated the database. Some 540,000 requests were submitted.
Ms. Trudel reported they broke down the budget question, where people indicated what they thought implementing HIPAA would cost. Some 196,000 said less than $10,000. Another 110,000 indicated between $10,000 and $100,000. About 27,000 said between $100,000 and $500,000. Some 8,000 anticipated between $500,000 and $1,000,000. And about 11,000 said over $1,000,000. Another 189,000 didn’t know. Most were spending less than $10,000, $10,000 to $100,000, or didn’t know.
Ms. Trudel invited questions and issues that needed to be researched. Asked if there’d been an analysis to determine if the 11,000 people who’d estimated over a million dollars represented 11,000 institutions, Ms. Trudel said they would do a breakout and make an additional cut to get better feedback on the size of the institution, type of practice, and expected implementation costs. Ms. Trudel said she’d check on whether they could do a refinement to tell if they’d done an additional un-duplication on either the name of the organization or authorized party.
She also reported asking for a tabulation of the number of covered entities who said they were using or would use a contractor or vendor. 158,000 said yes. 65,000 replied no. About 215,000 were undecided. Responding to the Subcommittee’s concern about the 48 percent who didn’t know if they’d need help, Ms. Trudel said she’d find out how many in the undecided section were also “I don’t knows” in the budget category and provide a breakout of undecideds in terms of when the requests were filed. She pointed out that the situation was different for undecided providers filing last May and those who keyed undecided on October 15. Dr. Steindel also requested a breakdown by demographic category. Members discussed that it would be hard to separate out true versus incremental costs. Ms. Trudel agreed, noting CMS had the same discussions about whether something was EDI or HIPAA.
Ms. Trudel said she’d next asked for a breakout on the reasons for delay. Some 250,000 said they needed more information about the standards. Another 254,000 were waiting for a vendor. About 368,000 needed more time. Another 226,000 still had to test. And 185,000 were waiting for addenda, and needed additional clarification on the standards. Only 51,000 reported problems completing additional data requirements. And just 36,000 cited problems implementing new clinical code sets. Ms. Trudel said much of that probably had to do with local codes. Money, staff, and hardware were fairly low, less than 100,000 for each. There were only 46,000 “others.”
Ms. Trudel reported they’d also done a distribution of the month and year testing would start and the month and year it would be completed, which raised questions about the data. About 20 covered entities reported starting testing months before the HIPAA legislation was passed. Five providers said they started eight months early. Members noted that those responders might already have been using X-12 standards or there might be data entry errors. Ms. Trudel noted the number beginning to test steadily picked up in December of 2000 and by January was doing well. They had 9,000 in April, 10,000 in August, 20,000 in September, and 39,000 in October. April 2002 spiked to 211,000. Noting that the majority of respondents would be compliant in time, Ms. Trudel said she’d do a re-run as percentages of the total, which would provide a better view. Mr. Augustine asked her to also re-run them as cross tabs, because these major categories would provide information.
Ms. Ward discussed a joint effort underway between WEDI and the Council for Affordable Quality Healthcare (CAQH) to come up with best practices to supplement the X-12N implementation guides. They also developed a template so, hopefully, all covered entities, particularly health plans, would issue their testing and implementation schedules on a central Web site they supported. Both were also working on a common credentialing format for all payers. Ms. Ward said they hoped to have the joint WEDI/CAQH initiative ready by January and encouraged the Committee to endorse it as best practice. Dr. Cohn suggested that Jim Shuping include this in his general discussion and that the Committee or Subcommittee review Web sites. Ms. Ward said she’d coordinate it.
Members discussed testimony they needed to solicit for the January meeting; providing guidance or best practices for those waiting for vendors and clearinghouses that weren’t ready to work with an unmandated version, and reiterated the need to get the rule out. Dr. Cohn suggested talking to Dr. Zubeldia about meeting with the vendors to gain a better understanding on this issue. He noted that at the January hearings they’d discuss improving the HIPAA process and issues linked to being successful between then and October. Members addressed the need to publish before it was too late and considered whether there was time to develop links on the NCVHS Web site and a listing of sites featuring best practices. Ms. Bebee and Ms. Trudel will assemble the resources discussed, so the Subcommittee could look at them at the January hearing and something related to the publication could be posted soon on the Web site.
Ms. Trudel reported that the final rule was nearing the end of the review process and on their post publication schedule for the end of the month. Ms. Ward said she was confident and always had been opposed to any conversation suggesting this rule wouldn’t get out in time. But she added that if it didn’t, it wouldn’t just be a question of how they’d test, but what they could do to implement 4010. As the last two years of DSMO work demonstrated, 4010 didn’t work. Ms. Ward emphasized that the real concern was what they’d do if they backed up six months, then had the Congressional review period, and it didn’t get out in time to implement in October. Dr. Cohn noted the Subcommittee and full Committee sent several letters to the Secretary and previous Secretary at the beginning of the timeframe trying to get everything moving. He said they were hopeful that, by the end of the month, they’d have the addendum. If it wasn’t out by January, the Executive Committees needed to send off a letter, but Dr. Cohn hoped that wouldn’t be necessary.
Improving the HIPAA Process
· Jack Rovner; J.D.; Michael, Best & Friedrich, LLC
Mr. Rovner presented the Secretary’s Advisory Committee on Regulatory Reform’s work regarding the HIPAA transaction rules. He explained that 29 appointees representing all segments of the healthcare industry and consumer interests examined the public comments about HIPAA and brought their own experience to the issues. They’d held a hearing during the WEDI Denver meeting with presentations from CMS and WEDI. Based on that input, the subcommittee issued a report and four recommendations (one on process, three on the rules) adopted by the full Committee.
CMS should adopt a defined schedule, with dates certain and a predictable, annual cycle for the issuance of updates of code sets, data elements and standards. In the regulatory process, proposed changes had to be issued in the Federal Register. The process had a certain period, and Mr. Rovner said feedback indicated a 30-day public comment period. The advisory committee helped revise the recommendation, building in flexibility for the defined schedule, because of CMS’s legal and other constraints.
Two other recommendations on process were withdrawn for further study. One sought to determine if certain changes could be updated automatically by the standard setting organizations (e.g., ICD-9, CPT-4) or in another way that avoided the cumbersome, prolonged regulatory process, yet allowed a 60-or-90 day notice period with CMS, who handled the code sets, having the ability to veto. The American Medical Association and CMS were being consulted about the concept’s practicality. The other recommendation withdrawn addressed a frequently expressed concern about the lack of code sets for the introduction of new technology and certain industry groups (e.g., long-term care). Remarking that the statutory framework for updating regulations was long and cumbersome, Mr. Rovner noted concern about technology approved by FDA and CMS for Medicare reimbursement that lacked codes; he emphasized that the present approach of mapping to existing codes was onerous and unrealistic.
The other three recommendations concerned rules that lacked four essential elements (sender, receiver, content, and purpose requirements) that usually appeared in a standard, causing confusion, uncertainty, and controversy. The 837 claim required the sender be a healthcare provider, the receiver a health plan, the purpose paid for delivery of healthcare, and the content information needed in order to be paid. Mr. Rovner reported that two or three of the standards lacked all four. The claims status, referral certification, and authorization standard had neither a sender nor a receiver requirement, suggesting that anybody could send or receive it. There was also confusion over the coordination of benefits’ sender and receiver elements.
Mr. Rovner noted that the existing rules required that a transaction that met a definition of a standard and was sent within the same covered entity be a standard transaction. He explained that claims status and referral certification didn’t have a sender or receiver requirement, and so appeared to be an internal standard transaction (e.g., a computer inquiry from the customer service department of a health plan to the claims department about the claim status). Mr. Rovner said the only rationale they’d found was that CMS seemed concerned about multiple-function covered entities (e.g., a hospital with an HMO as a division) and what happened when appropriate standard transactions were used in what appeared to be an intra-corporate transaction. He pointed out that the privacy rules directly addressed that issue by defining multiple function covered entities. Mr. Rovner noted the considerable concern expressed about the meaning of “within the same covered entity,” especially in terms of complying with claims status, referral certification, and coordination of benefits.
He reported the committee made a controversial recommendation on clarifying the regulations with respect to direct data entry exceptions. Noting broad-based confusion about what that exception allowed and how it would work resulted in wasted investments and moves in directions that might defeat the achievement of industry standards, Mr. Rovner called for greater clarification of what “direct data entry” meant. Noting the advisory committee disbanded in November, Mr. Rovner expressed confidence that HHS would address their recommendations.
Mr. Rovner pointed out that the problem of rules coming out on a staggered basis hadn’t been formally addressed. He said it was difficult to understand how they could have standard transactions without identifiers to ensure people knew whom they talked with. He noted that the rules coming out of the Department weren’t consistent. The rules had been so staggered that they hadn’t allowed for planning and budgeting. He questioned whether the issue was that the industry couldn’t take uniform, adoptable positions or if a problem with the DDE exception was that, by the time the final rules came out (which was when the exception showed up), it wasn’t in the proposed rule--but everybody already had moved ahead with the technology. Mr. Rovner depicted it as an afterthought that didn’t tell enough for the industry to understand its significance. He urged the Subcommittee to make these processes predictable, transparent, and consistent.
Mr. Rovner noted that one of his subcommittee’s points was that, with the ASCA Act, the Congress authorized money for appropriation that it never appropriated. That challenge required prioritizing and Mr. Rovner said he wasn’t in a position to suggest what came first. Noting there was a strong political/public constituency behind the privacy rules, he expressed confidence that they’d be published and enforced. But he cautioned that there wasn’t a public constituency behind the transaction rules. He pointed out, in retrospect, that the issuance of the rules could have been aligned so that, even with delays, they came out closer together. The Subcommittee heard a lot about how the rules would change, and people questioned spending money now. And Mr. Rovner noted there were changes on the transaction rules that, hopefully, would be finalized that month.
Because the process hadn’t worked well, Mr. Rovner said the industry perceived that this wouldn’t happen or work and was reluctant to make the investment. He emphasized that it would have helped if the identifiers were issued at or about the time the proposed rules came out with the transaction rules. Mr. Rovner reiterated that the rules were needed as quickly as possible along with a predictable schedule and date certain, so everyone knew every year if and when a rule change took place.
Improving the HIPAA Process
· Maria Ward, Chair, DSMO Steering Committee
Ms. Ward recalled that she’d been part of a WEDI work group that considered similar issues. Their primary issue regarded a predictable schedule for implementing these transactions. Another unfinalized recommendation regarded how they processed the types of changes and the transactions evolution. One suggestion was for a triage system where the approval process was consistent with the magnitude of the change. A preliminary recommendation was for the Steering Committee to assess the magnitude of the change and recommend an appropriate approval process. For example, correcting errata typos in implementation guides wouldn’t require regulatory process, but could be updated and changed with the next version or publication of the standard specification. Ms. Ward suggested that the Steering Committee was the place to assess the magnitude of change and recommend a process to NCVHS. Although a triage step required more time before the regular DSMO process, Ms. Ward said changes could then be addressed in a more expeditious way.
The WEDI work group also decided not to have more than two versions of the standards in production simultaneously. Noting there would be some parallel between versions, the group had discussed the need to define structure and dates. Ms. Ward noted that the work group had proposed the recommendation for claims attachments to the National Committee in February of 1998 and turned a draft over to the Department six-months later. Ms. Ward said she didn’t expect to see a proposed rule before the second quarter. Adding time for the comment process, resolution of comments, the final rule and OMB clearance, she said it would be eight years. Ms. Ward stressed that the time factor was becoming problematic and a big concern.
Ms. Ward echoed Mr. Rovner’s comments: there was confusion around understanding the exception for direct data entry and the requirements and incentives. She asked if an incentive was provided for doing direct data entry versus batch EDI, and noted the industry’s confusion over whether it was supposed to do that under HIPAA.
Ms. Ward noted deciding who the senders and receivers were in particular transactions also created confusion. Reading the definition in the regulation of the rule regarding whether the exchange transaction was a covered transaction was challenging. Ms. Ward reiterated the importance of defining when these transactions would roll out in order to provide predictability to those who had to deal with that data and change it for future versions. She stressed that this was a big area of frustration for implementers.
Scope of Work for Cost/Benefit Analysis for Migration to ICD-10-CM and ICD-10-PCS - Subcommittee Discussion
Ms. Pickett reviewed and amended as follows the draft of the cost/benefit analysis that captured comments made at hearings earlier in the year and a discussion at the last full Committee meeting about issues and concerns that should be part of the scope of work.
Assess full range of implications on current users of ICD-9-CM, Volumes 1-2 (diagnoses) to implement ICD-10-CM including costs and benefits (quantifiable and non-quantifiable) of change.
· Identification of affected entities and degree to which they will be impacted
· Identify costs, including opportunity costs, associated with transition including but not limited to information system changes, rate negotiations, recalculation of reimbursement methodologies, training, changes to forms, etc.
· Timing of transition, including impact of timing options on costs and benefits, potential return on investment, and interaction with other major health information implementation tasks.
· Identify immediate and future costs and benefits of improved data for, but not limited to, patient safety, outcomes analyses, reimbursement, disease management, utilization review, health statistics.
· Affected entities, including NCHS, CMS and other third-party payers and providers should be contacted to provide input to the contractor about their own assessments of costs and benefits within their organizations.
Assess full range of implications of replacing current uses of ICD-9-CM, Volume 3 (procedures), with ICD-10-PCS in the hospital in-patient setting including costs and benefits, (quantifiable and non-quantifiable) of change.
· Affected entities, including NCHS, CMS and other third-party payers and providers should be contacted to provide input to the contractor about their own assessments of costs and benefits within their organizations.
Assess full range of implications required to implement ICD-10-CM and ICD-10-PCS simultaneously (vs. separately).
The contractor will:
Members discussed next steps. Dr. Yasnoff and Ms. Pickett would present a revised version the next morning, so the process might start the following week. Changes would be e-mailed to members and comments received through December 18. A contractor would be identified who would discuss their plan with them at the January meeting. The contractor would talk to many of the affected entities and organizations; others would have an opportunity to present their own cost/benefit analyses and reflect on this work. The end date for the report was a week before the May meeting. The Subcommittee would review the report then, and decide whether to make a recommendation to the full Committee for an NPRM or have additional testimony, thereby delaying any recommendation until the September meeting.
Preparation for Complementary and Alternative Medicine Hearings in January 2002
· Richard Nahin, Senior Advisor for Scientific Coordination and Outreach, National Center for Complementary and Alternative Medicine
Dr. Nahin explained that the National Center for Complementary and Alternative Medicine was one of the 27 Institutes and Centers of the National Institute of Health (NIH). The Center was created by an act of Congress in 1998 and the budget for fiscal year 2002 was $104 million. He noted the Center primarily funded clinical, rather than basic research, reflecting the fact that alternative medicine was available to and used by the public. From a health policy point of view, Dr. Nahin said it was more important to test things currently used than discover mechanisms of actions. However, he added that as clinical trials began, the Center also initiated basic science studies to develop mechanisms of action.
Some 150 clinical trials were in progress, ranging from small developmental pilot studies of 20-30 people to clinical trials involving more than 3,000 individuals. The Center studied everything involved with complementary and alternative medicine, from botanicals, vitamins and other dietary supplements through procedure-based interventions (e.g., acupuncture, chiropractic, massage) as well as energy type therapies (e.g., Reiki, Qi Gong, therapeutic touch). The Center also investigated placebos.
The Center collaborated with other institutes at NIH, looking at alternative medicine across all disease areas, and with the Agency for HealthCare Research and Quality (AHRQ) on evidence-based reviews and CDC on the National Health Surveys. Its first study, funded in 1999 and looking at St. John's Wort (Hypericum) for major depression, was published in JAMA this year. The average clinical trial took three-to-four years and another year-and-a-half to write and publish. Most studies were still ongoing, but Dr. Nahin predicted the Committee would see many of their papers within a few years.
Dr. Nahin reported that the Center developed an in-house system for coding grants, so they could produce reports for Congress and other federal requirements. Alternative medicine was divided into five domains: mind/body medicine; traditional systems of medicine (e.g., traditional Chinese medicine); biologics (including dietary supplements, botanicals, energy-based systems and manipulation); and other body-based systems (including massage and chiropractic/osteopathic manipulation). The Center cross-tabbed its alternative medicine coding using dimensional diagnostics (e.g., coding an acupuncture grant as acupuncture and cross-coding it amesis caused by chemotherapy).
He said the Center had a small health services research portfolio and a clinic providing alternative medicine could look at outcomes in a specific patient population. He believed that they were developing their own coding systems, based on needs of their clinic or practice. He noted some big providers (e.g., Blue Cross/Blue Shield, Group Health of Puget Sound) had large patient populations using alternative medicine and provided limited services.
Dr. Nahin explained that the Center used health services research and other epidemiological studies to set their research agenda. Six criteria shaped their agenda: (1) the public’s use of a given intervention; (2) specific disease being treated, giving greater weight to those associated with high mortality or chronic morbidity; (3) the quality and quantity of preliminary data; (4) availability, at least for clinical research, of patient populations willing to be randomized (in some clinical trials, people had a bias towards alternative medicine); (5) if investigators had the required scientific expertise; and (6) the broad definition of politics within both the academic center (whether it’s willing to allow studies) and the regulatory process for botanicals and dietary supplements (the Food and Drug Administration allowed dietary supplements that didn’t make clinical claims to be sold as foods; products that did were treated as drugs and involved a different process). Algorithms were created from those variables, as well as cost. Each year, the Center prioritized what had to be done and issued initiatives for RFAs, grants and contracts.
Asked if the Center was in the early stages of categorizing these treatments and if more scientific evidence had to be brought to bear before they could categorize diagnoses and treatments differently than currently under ICD and CPT, Dr. Nahin said practitioners of a traditional system of medicine diagnosed in different ways than the conventional physician and had different diagnostic categories. He noted that one thing the Center looked for was correlation in diagnostic schemes for certain diseases in which conventional diagnosis systems weren’t accurate (e.g., irritable bowel syndrome, fibromyalgia) that were really a diagnosis of the absence of anything else. Dr. Nahin said an alternative medicine diagnosis overlay might allow producing more homogeneous groups of patients.
But Dr. Nahin said conventional diagnosis probably would have to be used for insurance claims related to alternative medicine practitioners, even though it didn’t match how patients were diagnosed. For instance, a traditional Chinese medicine doctor was interested in the balance of chi within the body and might consider someone with osteoarthritis to have a stagnation of chi going to their liver. But chi was difficult to measure and lacked a coding system. Realizing the insurance company wouldn’t know how to deal with it, the alternative medicine practitioner would probably use osteoarthritis as the diagnosis.
Dr. Nahin said a Harvard University study stated 42 percent (CDC in the National Health Interview Survey cited 22-25 percent) of the U.S. public used some form of alternative medicine, at an annual cost of 13 billion dollars, mostly out of pocket. Dr. Nahin pointed out that AHRQ’s Medical Expenditure study (MEPS) that only looked at practitioner use, not self-medication, came up with a much lower figure.
He noted that in many European countries (e.g., Germany, Italy, France) alternative medicine was part of standard care. In Germany, physicians were required to take courses on botanical products and many used them and homeopathy in their standard practice. Acupuncture was part of the National Health Plan in France. Noting that Europeans must have much data within their insurance systems, Dr. Nahin suggested that the Committee identify people from those countries to testify.
Dr. Nahin recommended that the Subcommittee talk to groups on the West Coast and Oxford Health in Massachusetts that currently reimbursed to some degree for different alternative medicine procedures. Although Blue Cross/Blue Shield didn’t feed into the CMS coding system, he suggested they had a broad patient base and would have worked out many of the problems. Dr. Nahin provided the Subcommittee with a list of individuals from the four predominant alternative medicine practitioner groups that tended to be licensed and certified in the most states, had the most structured education programs, and were most involved in research. He said he’d also submit a list of HMO’s and third-party payers reimbursing and a list of people doing outcomes research in this area.
Members proposed two provider panels (one comprised of individuals specializing in alternative medicine; another of complementary medicine practitioners with both a medical degree and alternative medicine credential) and a third panel of payers and outcomes researchers.
Ms. Molina, CEO of Alternative Link, said from their perspective the issue was integrative healthcare, which included conventional healthcare nursing services, alternative medicine, alternative healthcare, complementary healthcare, and spiritual interventions. Ms. Molina noted that Alternative Link had done an assessment of the complementary and alternative medicine arena and compiled statistics categorized different healthcare stakeholders and what this area of care meant to each. In considering panelists, Ms. Molina suggested three categories of uses for coding in complementary and alternative medicine: (1) research (e.g., scientific, health services based, clinical trials); (2) supply chain management--on the product side tracking product flow from a manufacturer, grower or botanical through to the end-user; on the services side everything from collecting data on diagnostic, disease prevalence and incidence, to actuarial data within an insurance plan in order to construct health benefit plan design or do infrastructure activities (e.g., health benefit plan design, managed care and provider contracting, utilization management functions, claims management functions); and (3) business transactions.
Alternative Link identified users of codes as: billers and coders, claims management services, practice management systems and companies, third-party administrators, administrative service organizations, self insured employers, health plans, integrative healthcare clinics, independent CAM and nursing practitioners, insurers, relative-value studies, incorporated or RUT-type experts, centers of excellence, academic research centers and schools.
Noting some minority health experts had expressed interest in the code set, Ms. Molina said one way to categorize was based on the importance of the code to pressing public health matters like disparities in minority or ethnic healthcare. She added that knowledge of all the practitioner types available to the general public was also of national security significance in terms of bioterrorist response.
Ms. Molina noted they’d already heard about different ways to categorize the practices, whether they were mind/body traditional systems, manipulation, body- or energy-based biologics, as well as the different practitioner types. She said another significant area in assessing both the market and the codes needed was the dynamics that would influence different players’ perspectives on capturing data around coding. Noting that key issues were economic and political as well as social or moral/ethical incentives, Ms. Molina advised determining whether the perspectives of the people presenting on panels were based on economical or political experiences or issues of public health. She said having a way to differentiate would help the Committee form good decisions.
Ms. Molina emphasized that financial incentives inherent in the current coding systems enabled some practitioners to be paid more than others. While Ms. Molina thought a lot of practitioner groups recognized the public health benefits of specific coding for their services or ways of identifying them as the caregivers, she said many were also afraid that they’d be financially or politically discriminated against if there wasn’t more precise coding for what they did. Ms. Molina offered the Committee Alternative Link’s presentations on: the market, health policy issues and dynamics, and what historically contributed to public health and financial challenges.
Ms. Bickford pointed out that the National Library of Medicine (NLM) grappled with inclusion of the complementary and alternative medicine languages, therapies and outcomes in their literature, and suggested that the Subcommittee ask how they dealt with that. She encouraged everyone to look at the world from a broader view than the pathophysiology medical model, so they didn’t close out their thinking and evaluation of the environment and what needed to be accomplished. Clearly they had to look at other clinicians and practitioners in the integrative healthcare environment and she suggested they include the healthcare consumer. Web products available for vendors promoting Internet-based healthcare information were another area she suggested.
Ms. Gianini suggested that a way to put one’s arms around alternative medicine was to focus on what the states licensed as healthcare services. Asked who’d be excluded, she said there was no state license for reflexologists, but one could have a code for an M.D. who took a Reiki course. Reiki could be researched effectively by starting research from a core or plenary license, looking at Reiki through the practice of a health practitioner (e.g., massage therapist, M.D., or chiropractor). Dr. Nahin noted that the four practitioner groups in his list probably were 99 percent licensed practitioners.
Ms. Gianini remarked that one of the challenges about how limited or expanded code sets should be was that everyone lived in a managed care environment where contracting for services was based on predetermined fees for services. Limiting or expanding coding in the administrative billing area, one had to be cognizant that in order to negotiate and pay fees for services, the coding had to be specific enough to get down to an appropriate fee for the service provided on the administrative side. And, on the PMRI side, it had to be specific enough to convey information indicating new ways to do things that saved money or could be more effective.
Recalling Ms. Gianini’s guidance that they focus on areas with licensed practitioners, Mr. Blair asked how fast the healthcare domain was changing and if homeopathy (currently licensed by only one-or-two states) should be included. Ms. Gianini said the priority of the person using the code set determined that. She noted the codes were developed to provide insurance companies with a tool that fit the set used for conventional medicine to relay information needed for NCQA certifications, managed care contracting, capturing data, and looking as an underwriter at information coming in, so premiums could be priced appropriately. Ms. Gianini said if one state licensed a practitioner, that individual was considered a healthcare practitioner.
Ms. Molina said Dr. Nahin’s list of providers for alternative medicine was appropriate, based on licensed practitioners, but incomplete based on scientific evaluation of emerging technologies. With a focus on administrative simplification and transactions, she agreed that the highest priority practitioners were licensed healthcare providers in a specific state. But she said in creating an infrastructure for studying the delivery of healthcare and approaches that generated the most significant economic and health outcomes, one would bring in a broad range of practitioners. From a scientific perspective, the goal was to get a code on every product, so you could evaluate demand and support consumer choice and supply chain management.
Observing that health services researchers looked at claims data and, if they weren’t included in the process, it wouldn’t reflect their needs and how they folded into Administrative Simplification, members noted that they wanted to remember research and other needs.
Ms. Bebee noted they had a list of providers for alternative medicine and Dr. Nahin would supply lists of health plans and researchers. Ms. Trudel and Dr. Nahin will solicit other staff volunteers, schedule a conference call with Dr. Huff and Dr. Zubeldia to develop the list in detail, and draft questions. Dr. Cohn noted that one day of the January 29-30 hearing was set aside for three panels on this issue.
DAY TWO Mr. Blair said the Subcommittee would begin with a report on the CHI Project. The rest of the morning members would explore the framework for PMRI terminologies, as they prepared to work toward some consensus on their criteria. He noted this effort followed the HIPAA directive that the Committee study uniform data standards for PMRI and electronic exchange of that information, and make recommendations to the Secretary. Initial recommendations to the Secretary in August of 2000 gave the guiding principles for selecting PMRI standards and a set of recommendations for accelerating development of PMRI standards and implementation. In February of 2002 the Subcommittee produced the first phase of PMRI standard recommendations for PMRI message format standards. And in August, they began exploration leading to recommendations for PMRI terminologies. The day’s goal was deciding on the criteria for selection. The first step was agreement on the framework or categories of PMRI terminologies.
The Subcommittee intended to send out the questionnaire seeking information about the criteria to the terminology developers by January 10. Five weeks were allotted for return of the questionnaires and another five weeks for the consultant, Dr. Sujansky, to compile and follow-up on the information. Dr. Sujansky will report to the Subcommittee at the March 24-25 meeting. A hearing for users of PMRI terminologies, including vendors and providers, is scheduled for May 21-22. Time will be available in June and July for additional expert testimony necessary for clarification, so members could develop by September an initial draft of recommendations to the Secretary on PMRI terminologies. Report on the Consolidated Health Informatics Initiative, Steven Steindel, CDC Dr. Steindel reported that the CHI Project had been working intensively on a parallel path for identification of terminologies with standards. He shared a document describing the results of a work group charged with looking at various terminology domains of importance to electronic health records and prioritizing them in terms of how quickly work groups could be established in those areas and present recommendations for terminology. The group looked at CHI’s internally generated list of domains used for a budget, work from Connecting for Health, and notes from the Subcommittee’s August hearings. Dr. Steindel said terminology was a problem in using these various sources, because similar domains were referred to by different names and there was looseness in interpreting whether they were the same. Based on these sources, four priority domains were developed to meet this charge, which was different than NCVHS’. The first priority were domains with reasonably apparent terminologies, so work groups could select appropriate terminologies to recommend. Dr. Steindel noted CHI’s goal was the interchange of healthcare data among federal agencies, which cast a different scope than what the Subcommittee looked at. The second priority was domains very important to the electronic medical record with terminologies that weren’t as mature. The third priority was domains where terminologies needed extensive work or didn’t exist. The last priority included terminologies essential for the electronic health record that were maintained by other groups and should be adopted as federal interchange standards (e.g., HIPAA). Sub-domains were listed for information purposes. Dr. Steindel said one charge the work group realized it couldn’t do was to exhaustively define the domains. Terminologies or their charges overlapped with other work groups and were hard to specifically define. CHI hoped the final results would be a similar set of terminologies, offering the Council options in which components would be most appropriate. Dr. Steindel reported that the first priority listed demographics, diagnosis and problem lists, interventions and procedures, laboratory result contents, medications, and text based reports. Laboratory result names weren’t mentioned because bids and an approved CHI recommendation identified lab LOINC for that purpose. Encounters, history and physical, immunizations, and population health comprised the second priority. Genes and proteins, multi-media, various wave-form descriptions, nursing, physiology and supplies made up the third priority. Dr. Steindel said he was pleased that Dr. Sujansky included the latter, because it usually wasn’t considered and was very important. Billing was listed as a category under terminologies used by other processes. HIPAA was included. Chemicals, disability, scientific and fundamental units were listed. Work groups formed in six terminology domains (demographics, interventions and procedures, laboratory result content, medications, encounters, and immunizations) start work early in 2003. Asked what was meant by encounters, Dr. Steindel said CHI posed the same query. Dr. Cohn considered an encounter as diagnosis or intervention. Noting others’ reactions were similar, Ms. Humphreys encouraged the Subcommittee to consider how they’d define it. Dr. Steindel said encounters came to mind because they were a prominent feature in the billing transactions. CHI made it a domain because the clinical encounter was different than the one in billing. They’d put it in priority two so it could be further defined. Dr. Steindel added that it was one of the first domains selected as a charge for a work group because CHI believed definitions coming from encounters would help drive the definitions and terminologies used for diagnosis, history and physical. Dr. Huff reported that HL7 did a lot of work on this and couldn't get anywhere with transactions until it distinguished between a billing encounter versus multiple face-to-face interactions with clinicians at a given facility, and people seen at different sites on the same campus or on the telephone or Internet. Ms. Humphreys reported that the first thing CHI did was to seek approval in the federal agencies for the Committee’s February 20 recommendations to the Secretary. She said NCVHS would want to know the work groups’ plans and timelines. Noting the Subcommittee and full Committee acted as CHI’s advisor, Dr. Steindel recommended that CHI seek advice on domains the groups selected. Members agreed that the two groups could share a very synergistic back-and-forth relationship. Dr. Steindel anticipated a three-month process for the first priority area. In other areas the groups might have to define domains before they could look at terminologies. Dr. Steindel said a timeframe leading to recommendations by September was synergistic and met their needs. Dr. Cohn reflected that much of the value of what the Subcommittee tried to do was in setting up a framework and sense of how things fit together and requirements. He considered the report a valuable additional cut with a specificity that might “put meat on the bones.” Ms. Greenberg said any organized way of gaining input for CHI from the private sector would be done through the Subcommittee as part of the interactive plan. Discussion on Scope and Criteria for PMRI Terminologies - Committee Discussion Members discussed using Dr. Sujansky’s working document to guide the Committee toward consensus on the criteria for selection and folding that into a questionnaire as the first part of their information gathering activities. They noted the idea might be modified as they went along. Members noted they needed general consensus on how things related and the major categories before they could gather criteria. Mr. Blair proposed drafting a questionnaire the Subcommittee could look at the first week in January. Noting Dr. Sujansky had pulled together the members’ comments, Mr. Blair suggested he step them through the document. Before he began, members reviewed the sentence in the law specifying that the Committee should study issues related to the adoption of uniform data standards for PMRI and the electronic exchange of that information. Dr. Sujansky reviewed an update of the second draft of the document, seeking agreement on a framework enabling the Subcommittee to consider the criteria. The first change, responding to comments that a single terminology wasn’t possible or practical, recharacterized the core as a small group of tightly integrated terminologies. Many had also commented that Dr. Campbell's original conceptualization (which had no hierarchical or layering relationship) wasn’t the best or most appropriate way to represent relationships among these terminologies. A change in nomenclature replaced the layer terminology with the notion of groups. Numbering of the non-core sets also changed, based on feedback that the HIPAA and administrative terminologies had precedence in terms of the project’s goals. Terminologies in Layer 3 became Group 2. Dr. Sujansky further defined and characterized legacy, clinical terminologies now in Group 3 and motivated the need for them. Other changes included revising the content of section 4.2 (Domain Coverage of the Core Terminologies) and removing the list of preliminary candidates, based on comments that the list was premature. Dr. Sujansky noted they could decide on criteria and generate a list of candidates after the survey. An initial list of domains for the core terminology group to cover was added to guide the selection process and initiate thinking about terminologies required. Dr. Sujansky noted that Ms. Humphreys’ suggestions about an appropriate role for NLM’s Unified Medical Language System (UMLS) were built into a new section. Other additions included the notion that hierarchies in the core terminology group probably wouldn’t be adequate for all data analysis needs, and acknowledgment that the model had to accommodate local, customized or task-specific hierarchical relationships. Dr. Sujansky said the Subcommittee and full Committee would recommend an initial standard terminology architecture. The current proposal described in the document included: the composition of the three terminology groups, the degree of integration of core terminologies, the relationship of core terminologies to the other groups, the relationship between the recommended terminology and message standards, why this relationship was important, and an organizational model for governance, licensing and ongoing maintenance of the core terminologies. Ms. Humphreys credited Dr. Sujansky with doing a good job of addressing everyone’s comments. Dr. McDonald said it was an improvement, but he cautioned about making judgments when establishing criteria with Groups 2 and 3. He thought Group 2 worked easier, because it described and characterized the terminologies, but he found Group 3’s circular logic confusing. Mr. Blair suggested the selection of terminologies could be used as a tool. He said they might be able to make selections or recommendations in the core and Group 3. But anticipating many situations where the terminologies didn’t meet their criteria, he suggested that recommendations could be made to terminology developers, so they could refine the architecture in the future. Mr. Blair emphasized that this might even be more important than their initial selections. Dr. Huff approved of how Dr. Sujansky defined the purpose, users and idea of a core, though he doubted they’d gain anything categorizing Groups 2 and 3. If they decided to map between the core and CPT-4 and ICD-9, Dr. Huff suggested they explicitly delineate everything they wanted to enable and not worry about groups. Dr. McDonald agreed, cautioning that groupings could create friction. Mr. Blair expressed concern that coming up with criteria for the core might make this more rigorous and demanding, because of the functionality and how the core terminology would be used. Noting it probably would be inappropriate to use the same criteria for Category 3; he cautioned that if they didn’t come up with a separate set they’d send the message that whatever people had was fine. People would interpret that there was no need to improve clinical specificity, interoperability and mapping between terminologies.
Dr. Huff acknowledged that they needed criteria for the core. But he pointed out that other things they were mapping weren’t justified by their structure or content, but because they’d been mandated for other business uses and/or were so entrenched that they couldn’t be changed. Given strong criteria for the core, he said there was a need to define these other mappings and relationships in order to make the core more useful. Dr. Huff noted ICD-9 codes were mandated, and criteria also weren’t needed for Groups 2 and 3.
Mr. Blair said his understanding was that Group 2 was basically HIPAA identified. He agreed the billing and reimbursement codes didn’t need selection criteria. But although other clinical-domain-specific terminologies (e.g., the nursing code sets and probably the IEEE device code sets) were specified as message formats, codes (e.g., DSM behavioral codes) were involved. Dr. McDonald pointed out that they didn’t know if some of those would end up in the core. Responding to Dr. McDonald question about why things were being recommended, Ms. Humphreys noted the Committee issued two reports. The first indicated they’d use a specified process and recommend patient medical record standards. The Committee hadn’t thought regulations were needed, but said there should be guidelines and the federal government should be early adopters. Last February, the first set of recommendations was sent in. Now it was time for the next set, which addressed vocabulary. Ms. Humphreys said it was clear they were doing this so the next set of recommendations could be sent in. She said the process seemed to work well because it looked like, soon, HHS, DOD and VA would adopt the first set of recommendations. She commended the Committee on their good work. Ms. Humphreys said she’d heard that the document should focus on the selection of the core and acknowledge the need for functional mappings to the mandated HIPAA code sets. There had to be some accommodation for mapping to terminologies that didn’t make it into the core and the fact that they might continue to be useful for local purposes. Ms. Humphreys suggested that the elevation of these other terminologies caused more problems than Dr. Sujansky’s document indicated.
Dr. McDonald advised against asserting things unnecessarily in doing their main job and recommended compromising between what Ms. Humphreys and Dr. Huff said to reduce text. Dr. Cohn noted one possibility was a survey that clarified whether an entity was Category 1 or 3. He advocated focusing on the core terminology and noted that some things had to occur between the HIPAA-mandated terminologies to make sure everything connected well. Dr. Cohn suggested there was a set of requirements, which had to do with good mapping, that other terminologies had to meet. Ms. Humphreys agreed that there’d be enough trouble selecting the core group without having to make nominations for three. She concurred that the issue was more that these things existed and something had to be done to go forward. Dr. Zubeldia noted that a number of terminologies in Group 3 and some in Group 2 were exclusive to their own domain and didn’t have mappings to other terminologies. Dr. Zubeldia said CDT-2, which was the only thing in its domain for administrative use, could be a core for administrative uses. Ms. Greenberg said she’d been thinking in that category of ICF, which was practically the only comprehensive classification of functional status. She, too, had issues with the original three layers. And she agreed that the Subcommittee should focus on the message standards and thought they were looking at the core terminologies to be used for primary documentation of clinical care. But she expressed concern about what she felt was a lack of appreciation regarding what classifications were used for that she didn’t want in the document. With so much clinically specific data for a number of purposes, better classifications were needed. Dr. Huff acknowledged that the mandate wasn’t arbitrary. The classifications and terminologies were essential and probably as good as could be imaged for their purpose, though that didn’t mean they were solely for that. Ms. Greenberg said she liked the language in the document delineating Group 2, which went beyond billing to include the need for mapping and classifications, because many people felt the Subcommittee would develop or name clinically specific vocabularies and wondered why they couldn’t forget about more updated classifications. Dr. Sujansky summarized their perspective and comments. The intent and goals of characterizing groups 2 and 3 wasn’t to select terminologies that helped meet the objectives of standardized terminologies for PMRI unto themselves. The assumption was that terminologies in other groups wouldn’t meet those objectives, but that the core group met objectives of interoperability and comparability that were the ultimate goals. The purpose of the other groups was to acknowledge other important terminologies that would continue to exist and had to be accommodated by whatever core terminology ultimately developed. Provisions were needed to help those using the other groups use and adopt core terminologies for clinically specific data capture and data analysis. Dr. Sujansky explained that the core terminology was intended to provide a new function. The other terminologies and relationships to them were intended to promote adoption of the core terminologies, so they could say how the new terminology recommendations related to the government’s decision last year that HIPAA-mandated codes needed to be used. HIPAA mandated terminologies would be acknowledged as important and, ideally, there would be mapping to those terminologies from the clinically specific terminology. It would all fit together, make sense, and work. Dr. Sujansky emphasized that was why Group 2 (i.e., the relationship of the core group to the other terminologies) was important Dr. Sujansky was amenable to dropping Group 3 as one they’d select member terminologies for. He suggested discussing a liaison-type function between developers of the core terminology and other important terminologies that weren’t in the core (e.g., the drug terminology, the commercial drug decisions support system). He emphasized acknowledging the importance of interoperating with some existing terminologies, so the core terminology could gain acceptance and adoption. A mechanism needed to be in place to promote that interoperability, but it didn’t need to include a selection of specific terminology within that group. Dr. Steindel said Ms. Humphreys and Dr. Sujansky summarized it nicely and pointed out the difficulties. From a theoretical point of view, he liked the idea of Group 3, but cautioned that it was difficult and shaky to characterize a terminology as Group 3 in a reproducible fashion. Dr. Steindel noted Dr. Sujansky summarized that there were two classes: the administrative terminologies (however they characterized them) and the remainder of the clinical terminologies (some existing for specific purposes). Dr. Steindel proposed combining those other terminologies into one group and discussing them together. Both had to interact with, map to, and maintain a relationship to the core, so what they did with the other terminologies was essentially the same. Dr. Steindel said they should identify this other group of terminologies that didn’t fit in the core and had the administrative terminologies as a subset, or what they called Layer 2. The other subset was terminologies they’d talked about as Group 3. He said they comprised one group because the function of what they did with other terminologies with respect to the core terminologies was the same. Dr. Cohn agreed with some of what Dr. Steindel said, but he pointed out the value in specifically identifying HIPAA-mandated terminologies and showing how they fit in. Dr. McDonald said he wanted it simple; if they strove for perfect, complete coverage they’d attract contention and wouldn’t get done, Dr. Cohn he said he had a problem, because there wasn’t a strict line between administrative and clinical terminologies, classifications and nomenclatures. When he practiced clinical medicine, ICD diagnoses sometimes helped him take care of his patient. Even Volume 3 procedures or NDC drug codes sometimes helped. He wondered if the HIPAA-mandated terminologies, classifications and nomenclatures were the real core and only needed extensions to make things clinically valid, useful and helpful. Dr. Cohn suggested the terminology might start with the mandated pieces and encircle the entire core. Dr. Huff said Dr. Cohn’s observations were extremely useful; patient care couldn’t take place and a drug couldn’t be prescribed by using those coding systems. For example, referencing coding books, asthma only existed if the patient was over 19 years old. Pediatric asthma had a different code. The coding handbooks added nuances by detailing when to use a particular code. All those nuances were integral to the meaning of those concepts, making them inappropriate for documentation of primary care or message exchange. Dr. Hoff reiterated that it was essential to know the relationship between the kind of asthma when documenting that a person had a restrictive airway disease, but that there was a difference between that and the particular asthma code assigned for purposes of ICD billing, DRG grouping, or morbidity and mortality statistics. Dr. Huff agreed they were core, in the sense that they were essential. But he emphasized that it was confusing to think about them as core or primary documentation of healthcare. Dr. Steindel agreed the discussion was useful, but noted Dr. Sujansky pointed out in the document that they were talking about an information model, not the terminology itself. The terminologies they were looking at to recommend as part of the core described clinical situations. Administrative terminologies were helpful in assessing clinical situations, but didn’t do that job.
Everyone agreed that they only had to define criteria for selection for the core and that, even though they wouldn’t select other terminologies, next summer they’d probably provide guidance or recommendations related to how UMLS would facilitate mapping to terminologies that weren’t in the core. Dr. Huff wondered if considering that there was a core clinical versus core administrative component made it clearer that they served a different purpose, but were mutually important. Noting ICD-9, CPT and HCPCS had been adopted, Dr. Zubeldia recommended simply saying they were core terminologies not yet adopted as standard under HIPAA. Mr. Blair observed that they weren’t considered clinically-specific in the sense the Subcommittee was using. Ms. Humphreys pointed out the need to establish that this activity was based on the previous report and recommendation leading to what might be adopted as guidelines, but not regulation. Dr. Zubeldia suggested simply saying their work wasn’t terminologies already adopted by the Secretary, but new and clearly for core terminologies. Mr. Blair said the law already gave them a mandate; they were looking at PMRI terminology for the core. Dr. Huff complimented Dr. Sujansky and said the important thing was to recognize the section on UMLS and it's potential role. He said the reasonable way to go forward was for the core terminologies to be a subset of the UMLS metathesaurus under NLM, which would be responsible for making them usable for clinical purposes. Dr. Huff said their default position should call for the appropriation of additional funding and for NLM to support this purpose. Dr. Cohn noted the Subcommittee believed UMLS potentially was a low-cost distribution modality for terminologies and had already gone on record supporting that. Dr. Fitzmaurice said his only concern was incorporating this in a document about scope and criteria. He thought it might well be one of their recommendations, but that they should reach it rather than begin with it. Dr. Huff said he thought it important to have in the document because, if it was their stance, it’d be useful to have people comment on it. With public testimony, they could ask other questions about how easy it would be to incorporate people’s terminology in the UMLS metathesaurus and how it would integrate with other things people envisioned in the core. Participants defined scope as the domain coverage of the core terminology group. Noting that everybody meant something a bit different (e.g., CHI’s tabular list or the Subcommittee support of patient care) when talking about PMRI and scope, Dr. Cohn said he was perplexed about what would be in or out beyond use cases, how many use cases they’d create, and how they’d determine who should receive questionnaires. Mr. Blair suggested that the first-or-second cut on the scope was in Dr. Sujansky’s document. From that day’s discussion, he said it sounded as if the boundaries would be defined as they came to consensus on the criteria for core terminologies. Dr. Sujansky said he agreed with Mr. Blair, but also with Dr. Cohn about the need to avoid a circular definition. Noting the criteria included good domain coverage, as defined in its scope, he said that scope had to be defined before they could apply that criterion. Dr. McDonald reminded everyone that this was just the first round and cautioned that, if the scope was too well defined, people would say the Subcommittee didn’t want to hear what they were saying. Members remarked that if the Subcommittee agreed on the criteria and sent questionnaires to all candidates, the process of seeing who met the criteria would define who was within the core. Dr. Cohn basically agreed with Dr. McDonald, though he added that approach didn’t indicate what wasn’t yet ready, but was needed to complete the core. He noted that Dr. Steindel's activity began to parse out what needed additional work. Dr. Huff viewed the core as an evolving set of things that might be forever increasing. He said they wouldn’t be able to compile and prioritize a complete starting list, but could begin with what they had and get input from HL7, public health, clinical trials and others as part of a self-defining process. Dr. McDonald suggested explicitly stating in the document that this was a step-wise iterative process and not a one-time winner-take-all.
Addressing the possibility that some candidates selected for core terminologies didn’t meet all the criteria, because some weren’t appropriate, the Subcommittee sought consensus on whether each was useful, and if it was quintessential (without it, the terminology couldn’t meet the goals of interoperability and data comparability) or desirable. Dr. Sujansky informally defined the first criterion, concept orientation, as elements of the terminology that were coded concepts with possibly multiple synonymous text representations and hierarchical or definitional relationships to other coded concepts. The Subcommittee noted the value of concept orientation was that it supported explicit synonymy and enabled other desirable properties (e.g., meaningless identifiers, concept permanence). Everyone agreed that concept orientation was quintessential.
The second criterion discussed was that, when the meaning of a concept was changed or refined, a new coded concept was introduced and the retired code didn’t disappear but was annotated no longer useful. Dr. Sujansky noted the validity of this criterion was that it supported longitudinal data analysis across versions of the terminology.
Members noted comprehensive domain coverage wasn’t quintessential. With a good vocabulary, one could always expand coverage and domain.
Dr. Sujansky defined meaningless identifiers as unique codes used to identify concepts in the terminology unrelated to the meaning of the concepts or their locations in the concept hierarchy. The Subcommittee agreed that meaningless identifiers were very important (they enabled multi-hierarchies, unlimited number of siblings in some cases, and reclassification), but could be compensated by explicit hierarchical relationships. Members emphasized that, even though meaningless identifiers weren’t essential, permanent identifiers that carried meaning were quintessential.
The Subcommittee described non-redundancy as each concept being represented by just one code in the terminology. They said it was important because it prevented false negative query results and necessitated more complex queries, even if all the codes were known. Non-redundancy made data analysis reliable. It was highly desirable, but not quite quintessential.
Non-ambiguity was depicted as each code in the terminology having a unique meaning (i.e., the inverse of non-redundancy). Members considered that being able to represent classes of concepts was distinct from not being able to determine whether the person who used the code meant the patient took the test or the substance. Theoretically, with standardized context and information one could decipher the meaning, but members noted that it might be quite a while before all the elements were standardized. So long as the vocabulary, itself, had such an inherent level of ambiguity, non-ambiguity was quintessential
The Subcommittee noted that multi-hierarchies were valuable because they prevented redundancy when concepts required multiple classification and allowed local hierarchical additions. Noting that the capability to take any vocabulary and create one’s own multi-hierarchies already existed, members agreed that this characteristic was desirable, but not quintessential. Members noted that a richer and more extensive hierarchy was extremely desirable in the terminology, both because of the difficulty of maintaining the concept of non-redundancy in a vocabulary that lacked hierarchical arrangement and as a valuable element of the terminology unto itself that supported data analysis and aggregation. Noting LOINC was non-hierarchical, but still one could infer a hierarchy because of the fields, members agreed that, while desirable, hierarchies weren’t quintessential.
While acknowledging that context information (e.g., past medical or family history) clearly needed to be represented in the patient or electronic medical record and that they might work toward a controlled vocabulary for the primary contract context elements, members considered this the responsibility of data representation, not the terminology itself. They said representing context information was important as a requirement for either a messaging or electronic medical record system, but wasn’t relevant in this list.
Members concluded that formal concept definitions, description logic, and infrastructure for collaborative development were all desirable. They also decided that explicit version identifiers (the knowledge of which version of the terminology a particular concept first appeared in and when it was retired) reduced potential problems for everyone and were desirable, but not essential. Even when this core set of clinical content was labeled, identified and updated with the UMLS published quarterly, the discreet version identifier had to be in the UMLS going in or no one could update anything properly.
The Subcommittee determined that availability of methods and tools to ease local implementation, explicit change set (having the precise criteria within each new version), incorporation of subsequent updates, support for local customization, and subsequent merging with updated standard were desirable.
Members also decided that, while probably any terminology would have to be substantially revised and upgraded, mapping was an important part of the demographics needed for the terminologies. They agreed that it would be important to ask who validated the mapping and exactly for what purpose.
Dr. Sujansky explained that he’d divided the non-technical criteria into three sections: ownership and licensing (e.g., intellectual property, cost and provisions for distribution and redistribution); maintenance processes aside from the technical maintenance processes about concept permanence (e.g., how many people were involved in updating, how input was received and integrated and processed, how frequently the terminology was updated); and the criteria related to viability and funding (i.e., the grayer area of how they knew they’d be around in five years and still doing what they did).
Mr. Blair noted the criteria were usually on a scale; there wasn’t a “black-and-white” on any of the general criterion. Dr. Steindel approved of Dr. Sujanksy’s descriptions, noting it was enough that they could do a draft and look at specific questions. Mr. Blair suggested they consider pulling from their criteria for selection of PMRI standards. Five criteria were general and not mandatory, because they were inherited. The law stated preference went to an ANSI-accredited standard, and in the terminologies there was only one. Dr. Cohn clarified that it wasn’t a criteria for code sets, but only for the message format standards. Mr. Blair said what was important would be to ask about the process for managing the standard, including different codes: was it consensus-based or balloted, and what were the criteria for membership. Ms. Humphreys said the wording of the HIPAA Administration Simplification section divorcing this notion of the actual accredited consensus standard process from the code sets was deliberate. Rather than focusing on the standard aspects, this asked who could submit requirements, how they heard back and how every legitimate need had a fair hearing.
Mr. Blair added that the other items were indications of the degree of technology and vendor independence and that licensing and implementation was relatively low cost. Ms. Humphreys cautioned that if they had to climb too many mountains, people would see it as a problem, even if it were free. Mr. Blair said that was why he specified the acquisition cost, however acquired. The other piece was the cost of implementation: if one had to acquire software to support this particular code set, that should be captured.
Ms. Humphreys remarked that three issues were being addressed: technology independence, the actual cost of the item, and the level of administrative hassle. She added that some who’d negotiated for license agreements knew this could be endless. Dr. Cohn agreed that they had to know the cost of the license, but he said they also had to recognize there were many ways to deal with this, including the government negotiating national licenses. Members agreed they’d ask several questions to get at the whole concept.
Dr. Huff agreed with the list, predicting they’d have interesting discussions when they considered weighting them. He said they needed to be unencumbered, so people retained the creativity and ability to incorporate all this in knowledge bases, make new representations and hierarchies and derive useful by-products. Asked about putting a thin shell of tools around a terminology and reselling it, Dr. Huff said derivation should be allowed; the only thing the Subcommittee was trying to prevent was needless proliferation of slight modifications on the standard.
Members noted they’d ask general questions about market acceptance: how many users terminology developers had, how many licenses, and if they were implemented. Dr. Sujansky said this input and their discussion of the criteria (both technical and otherwise) would be reformulated as candidate questions for the survey. A list of questions would be circulated to Subcommittee members the following week. Comments were requested no later than that Friday. Mr. Blair noted they were bound by their meeting dates and it would be tight, but he said he’d fold everything into a draft questionnaire. Mr. Blair will also indicate the developers they intend to contact. Ms. Greenberg proposed posting a notice that the Subcommittee was selecting information on technologies and inviting anyone interested to contact them.
Review Revised Scope of Work Draft for cost/benefit analysis for migration to ICD-10-CM and ICD-10-PCS - Committee Discussion
Dr. Cohn led the members through a final reading of the draft cost/benefit study on the impact on the healthcare industry of moving to ICD-10-CM and ICD-10-PCS. He said they’d find the contractor. Dr. McDonald would send any wordsmith changes to Ms. Pickett. It was noted that a task order could be awarded quickly, but that it would still take a few weeks. A motion to accept the draft of the cost/benefit engagement was seconded with all in favor.
Dr. Cohn noted hearing dates were scheduled for January 29-30, March 25-26 and May 21-22. He emphasized that the next year would be important for the HIPAA administrative and financial transactions implementation. They would talk with WEDI and other groups about that on January 30. Dr. Cohn was hopeful they could include any issues or major initiatives that came up within the hearing data schedule. But he cautioned that if they were particularly concerned about any issues, they might have to schedule more hearings. Dr. Cohn hoped this all moved forward, but he said they needed to stand ready to provide leadership over the next ten months. Dr. Cohn adjourned the meeting at 12:30 p.m.
I hereby certify that, to the best of my knowledge, the foregoing
summary of minutes is accurate and complete.
/s/ 03/03/04
_________________________________________________
Chair Date