The Work Group on Computer-Based Patient Records of the National Committee on Vital and Health Statistics held hearings on December 8 and 9, 1998, at the Hubert H. Humphrey Building, Washington, D.C. Present:
The NCVHS Work Group on the Computer-Based Patient Record (CPR) held a two- day hearing on December 8-9, 1998, the first step in its process to study the issues and advise the Department on the adoption of uniform data standards for patient medical record information and the electronic exchange of such information. During the two days, they heard presentations from and talked with six panels. Members had been sent questions to address concerning the Work Group's role vis a vis the government, the optimal role of government in this area, impediments to standardization, and other issues.
These panelists encouraged the Work Group to look beyond the tasks it initially envisioned for itself. They offered their views on the proper role of government in respect to standards development for the CPR and the Committee's optimal advisory role; identified major barriers; and stressed the centrality of clinical applications. As summarized by Mr. Blair, their views are that the private sector, while imperfect and moving slowly, is progressing on message format standards and medical technology, and on a number of current standards activities. The government should take a more facilitative role, while at the same time viewing the issue more broadly in terms of health care information infrastructure, and also do more to encourage existing standards efforts.
Both Dr. Schyve and Dr. Schutt described their organizations' perspectives on development of the CPR. Ms. Webman emphasized the importance of developing data standards that incorporate behavioral mental health information. She also called attention to the deficiencies of the May 7 NPRM with respect to behavioral and mental health services and noted other needed changes in this area. In the discussion session, the panelists expressed their views about the areas on which to focus during the next three to seven years. They also offered a number of concrete suggestions for short-term improvements in information policy. As with the previous panel, this group supported a governmental role in pulling together the diverse standardization efforts of private sector groups.
Dr. Chin described the use of "de facto standards" in his organization and the benefits both to patients and in terms of the business case. Dr. Mattison discussed a number of problems and issues to be addressed from the perspective of an integrated delivery provider. He also identified the forces working for and against standardization and he suggested areas for regulatory influence. Ms. Narcisi asserted that no federal legislation requires implementation of medical record standards, and it is unclear whether HIPAA applies to a single, large system.
She stressed that the government should confine itself to facilitating a voluntary private effort with significant physician input.
Dr. Orient's remarks elaborated on her basic assertion that there should not be a nationwide standard for the electronic medical record because it cannot be assumed to be in everyone's interest. Dr. Spackman introduced the College of American Pathologists' initiative SNOMED. SNOMED-RT, to be premiered in June 1999, will be available as a foundation for the CPR. He stressed the importance of a government role in facilitating progress in this area through testing and funding. Like most other panelists, he called the Work Group's attention to areas outside the four priorities it has identified. He also introduced the theme of the importance of international cooperation.
In the discussion period, the group talked about the cost barriers to adoption of the CPR, among other topics.
Mr. Gardner described the handling of patient information in his hospital, which is highly automated with the support of physicians who have found it time-saving. He noted that problems still exist in terms of information quality, and that using highly computerized systems does not necessarily solve these issues.
Mr. Arges pointed out that providers generally find it difficult to obtain the financing to undertake a massive information system investment. He recommended that government health programs such as Medicare and Medicaid be actively engaged in defining a core set of data standards, and that a national approach preempt disparate state approaches.
Mr. Bozeman described his organization and its CPR efforts over the past 15 years. Patients have common patient identifiers and use magnetic stripe patient identification cards. He and Mr. Arges agreed that federal funding cuts are impeding forward movement on standards development.
All members of this panel favored a strong governmental role in facilitating standards development. They offered very specific recommendations for the types of changes needed and where the federal government might show leadership. Dr. Russler drew an analogy to the FCC role in the HDTV process and said the government should participate in the development of standards on health care data communication to the same degree as the FCC has done in its domain.
There was support for Mr. Tonks' comment that an important factor inhibiting development of uniform standards is a lack of understanding of the data term or code placed in each data field of a message. This impedes plug-and-play interoperability. He suggested that government support would be valuable in this area. Dr. Tang's recommendation that standards be mandated to provide a business incentive was also supported by some other panelists, as was Mr. Tonks' idea of a terminology service bureau managed by a trusted agent.
In the discussion period, the group considered the government role in creating positive incentives. Both Mr. Mayes and Dr. Fitzmaurice described projects of their agencies that were well received by the panelists. Dr. Cohn asked that the panelists enlist their legal departments to send to the Work Group additional information on issues related to state regulation.
The group also discussed the challenges of transition to the CPR for the average hospital. Dr. Lumpkin expressed interest in a site visit to a hospital with automated records. All panelists supported Ms. Fyffe's idea that medical schools offer a required course dealing with standards of documentation of medical records. It was suggested that this also be required for the chief medical officers of health care organizations.
The issue of incentives and how to create positive ones was a consistent theme. Panelists were asked to send the Work Group their thoughts about the tension between vendors' desire for regulations and health care providers' opposition to them. Another topic of interest was how to level the playing field so that institutions without vast resources can participate in decision-making about standards.
Dr. Morris supported computerization of records and standards development, but cautioned against standards that would interfere with natural clinical processes. He cited the Human Genome Project as a good model of collaboration. Dr. Peters argued for a strong, proactive government role in standardization and cautioned against a dominant private sector role. He asserted that physicians are receptive to standards. Dr. NcNamara focused his presentation on drug vocabularies, an area in which he claimed that significant rapid achievements could be realized with modest investment. Mr. Radtke called attention to confidentiality and security as priority issues and, like other panelists, called for flexibility in the standards.
Dr. Cohn framed the discussion period by asking the panelists to advise the Work Group on steps the government should take to foster the alignment of incentives to develop a uniform standard.
Among other things, the panelists stressed the potential role of government, as a payor, and large employers in leading the way toward adoption of standards. They identified negative market forces that might wither as a result of judicious federal intervention. Dr. Peters agreed to compile a paper for the Work Group on his suggestion that a focus on codes be minimized. Panelists called attention to potential patient uses of data and electronic interaction between patients and providers.
Future dates for hearings include March 29-30 and May 17-18, 1999. The Work Group identified potential areas upon which to focus discussion in the future. Several participants offered detailed proposals that collectively created a broad spectrum within which the Work Group will develop its work plan.
Dr. Yasnoff remarked on the "remarkable consensus" among most of the witnesses that government should do something, particularly to facilitate and speed up the existing standards processes to coordinate them.
The group discussed Mr. Blair's suggestion that the first step in the process should be coming to an understanding of the spectrum of terms "clinical data repository" or "computer- based patient record" or "clinical information system." Some concern was expressed that the Work Group would get bogged down in definitions at the expense of an issue focus. There was apparent agreement, however, on the need to clarify what patient medical record information is, and what parts of it are important.
Several participants mentioned the need to study and articulate the business case for the CPR.
There was considerable agreement on the need to continue asking the country more questions about how the government can be of help. Groups from which additional comment should be solicited include the standards development organizations and the Departments of Veterans Affairs and of Defense.
Ms. Fyffe identified four parts to the mission of the Work Group: studying (identifying) the issue, adoption of uniform data standards (must define term first), patient medical record information (must define), and electronic exchange of such information. The group seized on this framework as helpful in organizing its work plan.
Dr. Cohn convened the first of a series of hearings to be held by the Work Group on Computer-Based Patient Records of the National Committee on Vital and Health Statistics (NCVHS) pursuant to the mandate of the Health Insurance Portability and Accountability Act of 1996 (HIPAA) for the NCVHS to study the issues related to the adoption of uniform data standards for patient medical record information and the electronic exchange of such information. Following introductions of the participants, Dr. Cohn welcomed the first panel.
Mr. Waegemann stated that uniform data standards for patient medical record information should not be limited to coding systems or data standards, but should include all standards related to medical records (e.g., on uniform information content, uniform structure and format, etc.).
He identified two types of impediments to widespread implementation of uniform standards for patient medical record information: subject matter and interoperability. Subject matter impediments include inconsistencies in clinical documentation techniques, styles, and formats; lack of clinical specificity in terms of coding system terminologies; and lack of adherence to principles of documentation. Impediments related to interoperability are organizational in nature, stemming from the fact that professional involvement in the standards development process of individuals in the medical specialties is insufficient. In addition, Mr. Waegemann cited a lack of national legislation that enables electronic health records; some states do not permit keeping paperless electronic health records.
He recommended that the Committee define minimum requirements for electronic health records vis-a-vis confidentiality, security of indication, data integrity, retention, and permanence; recommend to Congress that it override inconsistent state laws on electronic signatures, retention, and permanence of patient records; and recognize an organization as the center for standards coordination.
The major problems are the barrier represented by state legislative impediments, development of a personal identification system acceptable to all players, and creation of uniform document routes. A uniform documentation standard and standards for interoperability would add the most value.
Mr. Waegemann took issue with the emphasis on the four areas listed in the Work Group's fifth question sent to participants. He stated that it is more important (1) to focus on national legislation that defines the minimum requirements for implementing paperless electronic health records that would supersede state laws, (2) to draw additional professional organizations into the standards process, and (3) to give the American National Standards Institute’s Health Informatics Standards Board (ANSI HISB) a mandate to accelerate and coordinate the standards process for electronic health records.
Dr. Korpman seconded Mr. Waegemann's remarks, underscoring the criticality of accuracy in health care records by noting that death may be the consequence of improper use of information. Dr. Korpman stated that uniform data standards should be construed in the broadest possible way, attending to such issues as integrity, accountability, and auditability.
He noted that although experts have been working on standards for a long time, the standards still are not very good. Part of the reason, he postulated, is that there has not yet been a focus on designing an infrastructure that permits support for the best patient care. He recommended developing a standard that defines what the broadly contextualized electronic records data model needs to be and how the pieces must fit together in a HIPAA-compliant manner.
Among the steps ahead are development of standards for broad-based, comprehensive architectures, for end-to-end data flows, and for protections for individual identifiable information. Dr. Korpman agreed with the Work Group's four focus areas, but stated that they warrant cursory attention; the major push should be to develop a standard that reflects an overall vision of the system.
Dr. Stead contended that an appropriate yardstick to measure the efficacy of the public and private sectors in development of uniform data standards for patient medical information is whether this effort has a reasonable chance of translating it into desired increases in health care quality, access, and affordability in the near or intermediate term. If the consensus is that these benefits will occur, then it makes sense to focus on the issues raised in the four focus areas suggested; if not, the Work Group should conduct a broader inquiry to identify the set of initiatives that can be completed on a widespread basis and have the desired effect.
Dr. Stead raised questions about the short-term usefulness of computer-based patient records (CPR) using uniform data standards because (1) tools are not yet in place to manage the large base of nonstandard, biological concepts involved in health care, and (2) each record requires significant data entry.
According to Dr. Stead, among the appropriate roles of government in this effort are the following: (1) to continue to support basic medical informatics research, including structures for representing data that make relationships between terms and concepts explicit, methods for data mining and limiting retrieval to context, work flow, change management, and human-computer interface issues; (2) to provide incentives for the use of electronic records and best practice; (3) to support demonstration projects for effective implementation strategies; (4) to devise a system for certification of product functionality; (5) to decide on a standard way of identifying people; and (6) to require payers to make contract terms and authorizations available in a standard way as profiles to be referenced by practice management and billing systems.
Dr. Stead referred to a paper by Bidgood et al. in the January 1999 issue of JAMA, which outlines the framework for representing image acquisition content for images that are themselves represented within the DICOM standard. He also recommended parallel development in three other areas: (1) tools, such as interlinguas and other data modeling techniques, to manage content-rich standards; (2) standard software architectures to support services; and (3) emerging e-commerce technologies to support patient (versus facility) ownership of the record. Dr. Stead briefly summarized Vanderbilt's efforts to construct a CPR system.
Mr. Quinn asserted that Congress is frustrated with the inefficiencies of computer science in helping the U.S. health care industry become more cost-effective. The needs of disparate customers have led health care information system (HIS) vendors to create a variety of systems to meet that demand.
The U.S. computer software applications industry has not been able to leverage its successes from other industries to the health care industry for the following reasons: (1) Health care needs are unique and do not allow for easy reuse of applications from other industries; (2) standard processes are lacking in all segments of the health care industry, a condition that poses a challenge for computer programmers and for their companies' business decision makers; (3) clinical health care data are extremely complex; (4) rapidly changing technology serves as a disincentive for large investment in computer programs; and (5) no single vendor has demonstrated the technology and the ability to support the CPR at a large health care provider or large group practice.
The role of government should be to increase the pace of the work undertaken by the private sector, primarily by funding projects. In addition, the government must come to grips with contradictions in standards, perhaps by instituting a process that is less consensus driven.
Mr. Quinn recommended using standards that work, including HL7, X12, and DICOM. In addition, the codes for use within these standards must be standardized and the process semi-standardized. Mr. Quinn asserted that common and inexpensive electronic exchange of health information will not be enabled without more standards development and implementation.
Dr. Korpman referred the Work Group to his written testimony for a list of identified issues to be handled, such as characterizing what represents the medical/legal record, for which there are broad discrepancies from state to state. Other issues include trust in the quality of the data, stewardship of the data, wholeness and integration, accountability, security, scalability, availability, and reliability. Issues also must be addressed concerning process integrity and knowledge base, personalization, localization, and definitions.
Mr. Waegemann described the evolution of standards development in electronic medical records and then identified additional areas that must be addressed in the creation of a legal system for a paperless CPR, including data integrity and integrity authentication, storage and future availability. Mr. Waegemann also referred the Work Group to his written testimony.
Dr. Stead asserted the possibility of determining the few things that can be done that are not stymied by variability. He stated that striving to "get every container of clinical information in an open format (such as ASCII, DICOM), labeled with the right set of meta- data" would be a "giant step forward," and that this could be accomplished across the country within three years.
Mr. Quinn reiterated that industry's inability to leverage the software business model into health care is a significant barrier to success, and that work must be done over the long term to reconcile this issue.
Dr. Stead enumerated two possible pernicious legislative provisions: legislation that does not reflect an understanding of the requirements of a business's work flow, and a requirement that lacks the information base or standards to back it up.
Dr. Korpman suggested that in order to move the health care industry more quickly to CPRs, the government must establish an infrastructure with stringent requirements that must be met before payments are made, based on a system oriented to patient care that assures quality of data and usefulness for care.
Mr. Blair summarized the panel's presentation and sought comment. The private sector, while imperfect and moving slowly, is progressing on message format standards and medical technology, and on a number of current standards activities. The government should take a more facilitative role, while at the same time viewing the issue more broadly in terms of health care information infrastructure, and also do more to encourage the existing standards efforts.
Mr. Waegemann added that the transition from a paper system to paperless may require legislation providing that any clinical hospital can have paperless electronic health records if they comply with certain selected standards, standards that already exist. Dr. Stead suggested that a "political win" in the development of CPR over the next three to five years will promote the possibility of securing sustainable support to do what is necessary over the next ten to fifteen years. Instead of mandating processes that would require more overhead and more systems, Dr. Stead suggested that a few effective processes be selected, and that health organizations be helped to use those processes to use the data already in the systems to drive out the unnecessary variability in process.
Dr. Korpman suggested that the Work Group hold a separate discussion on the sequencing of standards possible to implement in the near term. Mr. Quinn reiterated that process variability is an inhibitor to enabling information systems technology in the health care industry, and he cautioned against assuming that the information technology industry can do this job at this time.
In response to a question from Dr. McDonald, Dr. Korpman asserted that the government can put in place high-level guidance on how people sort their data. Dr. Stead focused on the needs to solve the patient identity problem and to make any "container" of information accessible outside its system, and he cautioned against embedding business rules in the use of the data field. Dr. Stead stated the need for a meta-data concept model, wherein a term has only one meaning. Dr. Korpman expanded on the issue of concept representation. He suggested inclusion of a data type called "encounter," but that each organization establish as many examples as they can. But all data about that encounter must be traceable to the originator, authenticated, and end-to-end processed in the system.
Mr. Waegemann cautioned that detailed standard actions not be the focus of this discussion and that ongoing work in the standards organizations not be duplicated. He recommended three areas in which government initiative may be helpful: (1) Enabling national legislation to replace state laws, to describe the minimum requirements for data integrity issues, and to define principles of documentation in the electronic environment. These standards exist, and it is necessary that they determine the national minimum requirement for CPR systems. (2) Empowerment of standards organizations to coordinate issues such as ANSI HISB. (3) Use of NCVHS's clout to bring the professional organizations of medical specialties to the table who have not yet contributed to the dialogue.
Dr. Fitzmaurice asked the panelists to describe a useful model to explain the issues to a lay person. Dr. Stead advocated that by using existing tools and developing a system that represents the best of the "low-hanging fruit," significant improvements may be reaped for the health care system. He noted that the basic work has not been done to create, for the health care system, the equivalent of the genetic map under development by the Human Genome Project. He stated, "If we could have data about what we do, together with long-term outcomes, the computer would provide a time-lapse camera that would let us see which of those process variables made a difference and which did not." Mr. Quinn described the modeling process from an engineering perspective, stating that it is an aid to communication. Mr. Waegemann cautioned that modeling requires significant time and funding.
Dr. Schyve told the Work Group that his organization's accreditation manual enumerates information management standards that were drafted to facilitate and encourage computer-based management of personal health information. He described four standards relevant to the Work Group, and stated that three of those standards include the phrase "whenever possible," which reflects the general absence of universally recognized standards for personal health information content, format, and data elements.
Dr. Schyve referred in his written testimony to six factors that have hampered the adoption of uniform standards, none of which, he asserted, can be overcome by the public or the private sector acting alone. He asserted that overcoming these barriers depends ultimately on the value of the uniform standards and of the CPR.
Dr. Schyve presented his commission's perspectives on the value of uniform standards and the CPR vis-a-vis the following issues: (1) roles of the standards in CPR in improving care of the individual patient, particularly by multiple providers; (2) role of standards in continuous improvement of the effectiveness and efficiency of health care delivery; (3) role in assisting consumers and purchasers in value-based choices; and (4) role in supporting the U.S. quality oversight system. Dr. Schyve asserted that the Joint Commission advocates the adoption and implementation of uniform standards for health information and the CPR because it would facilitate the accreditation process, but more importantly because it would improve the care provided to the public.
Dr. Schutt stated that the mission of his company is to apply health care information strategically to improve the value of health care. The group's clients want to demonstrate that they are doing something worthwhile for their patients; to justify to their CEOs that their information systems provide value; to measure cost, use, and quality in medical care and productivity; and to compare their performance with like organizations.
Dr. Schutt explained MEDSTAT's system for translating clinical findings into ICD-9 codes, which permits the electronic classification of patients by disease severity, but with the loss of descriptiveness. MEDSTAT now uses claims data to determine electronically the services provided to a patient or group of patients; the economics of development will determine the possibility in the future of including such data as lab test values in such a database.
In his responses to the series of Work Group questions, Dr. Schutt noted the following: (1) Congress is seeking guidance on who will lead the standardization effort. (2) Barriers to implementation include the absence of a universally accepted standard and a clear leader; concerns over confidentiality, privacy, and security; concerns over return on investment in information systems; disagreement on the degree of complexity in health care that must be captured electronically; and lack of consensus on medical nomenclature. (3) The role of government is to facilitate closure on a standard and to offer mandates or incentives that motivate the industry toward its use. Dr. Schutt held out MEDSTAT as the "embodiment of the business case." (4) The government should narrow its focus on the "doable," the quantifiable. (5) A MEDSTAT model for communicating the need for a CPR is that of a patient leaving the house without identification; when the patient moves from one health care setting to the next, there is "amnesia," which impacts on the value and quality of the health care that can be provided.
Ms. Webman stated that her group's primary mission is to create and promote equitable standards and guidelines for ownership, privacy, confidentiality, quality, and accessibility of behavioral health and human services records. Ms. Webman's presentation emphasized the importance of developing data standards that incorporate behavioral mental health information. She noted a communication from the Center for Mental Health Services, stating that the Notice of Proposed Rule Making (NPRM) of May 7, 1998, on data transaction standards is "seriously deficient with respect to essential information on functioning mentally ill persons required for enrollment data, and is totally nonresponsive with respect to encounter data on the classification of mental health services." Ms. Webman suggested that technical consultants from CMHS meet with HCFA and ASPE representatives to address the issues.
Ms. Webman noted that individuals with behavioral health care episodes often also have needs in special education, justice, and other services; a medical record with data would have enhanced value. Moreover, few data elements in the proposed rules overlap with mandated child welfare data enrollment sets, voluntary Mental Health Statistics Improvement Program data sets, or TANF enrollment encounter data sets.
Regarding privacy and confidentiality, Ms. Webman referred the Work Group to a handout that describes a prototype infrastructure that would connect human services, education, housing, and health care, plus information on ideal content.
In her responses to the series of Work Group questions, Ms. Webman noted the following: (1) Lack of dissemination and knowledge within the behavioral health and human services community, lack of consensus on behavioral health standards, and inadequate attention to privacy protection are barriers to development and implementation of standards. (2) The role of government should be (a) to create protections consistent with the Consumer Bill of Rights Bill, specifically mentioning the needs of the behavioral health community with regard to privacy and confidentiality; (b) to protect the public trust; and (c) to develop and implement a business plan for the health and human services industry that will guide technology development. (3) The business case should be created with a public-private partnership that allows for full private investment.
Ms. Webman, in response to a series of questions by Ms. Fyffe, stated that the May 7 NPRM generally mentions behavioral health at the code level, but not at the general data element level. Terms such as outpatient and inpatient are included, but no attention is paid to social functionality scales. She referred to a forthcoming electronic copy of testimony that itemizes these issues. Ms. Webman explained that the Medicaid billing system does not capture important services, such as wrap-around services, provided by behavioral mental health professionals; because these services are not captured, it is difficult to justify their value.
To a question by Dr. Fitzmaurice, Ms. Webman asserted that research on describing and quantifying services vis-a-vis outcomes would help to establish the value of the services; the government could play a significant role in funding such research. Ms. Fyffe noted that data sets for other areas, such as long-term health care, are at a similarly immature stage of development.
The panelists discussed their views of high-value areas on which to focus during the next three to seven years. Dr. Schyve suggested pushing toward more uniform national standardization of such items as the medical record format, resolving content issues, focusing on the "atomic level" in classification, clinicians using standardization of systems in a standard way (as is done in other high-performance, high-risk industries), but at all times "creating safety." In characterizing the difficulty of the task facing the Committee, he offered a new definition of reliability: the ability to identify the changes that require interaction in order to prevent unintended consequences--which must be reconciled with the computer-science definition of reliability as minimal variation.
Dr. Schutt discussed the importance of using well the information available, and focusing on patient-related process. He also suggested that including pharmacy and lab data in the mix will enhance the descriptive ability of electronic methodologies in order to get to best practice.
Ms. Webman pointed out the lack of a family code in both Medicaid and Medicare, the presence of which would eliminate excess costs related to unnecessary multiple case managers. A focus on tangible usefulness to the patient-consumer, who could use for their own benefit the information in the system, for example, to retrieve information about providers on an electronic smart card, before they are asked to reveal private information, may enhance acceptance of the system. Ms. Webman also described the burdensome, nonstandard documentation requirements to which behavioral health providers must adhere. She pointed out the successful efforts of some jurisdictions to create a business plan based on such public- sector values as family focus, community basis, cultural competence, community drivenness, and coordinated care, and then to build a technology plan.
Dr. Schutt explained the advantages of using electronic pharmacy and laboratory data. In response to a question from Mr. Blair, Dr. Schutt stated that his clients respond to a critique of their data by improving their documentation practices; that MEDSTAT works with other standards organizations to standardize documentation requirements; and that it may be possible for standards to proceed without government facilitation. Clients are motivated to improve their data based on their knowledge of its usefulness and also on economic grounds. The government can make a great contribution by channeling the multiple efforts by diverse standards groups.
Dr. Schyve asserted the criticality of making the business case for the practitioners, the people who will be most affected by standardization, and by securing buy-in from those people.
Ms. Webman suggested that supporting public purchasers with education and holding them accountable for purchasing for qualityÄnot just low bidÄis important. At present in the behavioral health field, one cannot document the cost of an episode of care.
In response to Dr. Cohn's question, Dr. Schutt described the manifestations and ramifications of poor data quality. One consideration is that coding records is done mainly with reimbursement in mind, not improvement in quality of care. Errors in the system relate to how what happened to the patient is captured and how the data are recovered.
Dr. McDonald raised the issue of measuring by inadequate standards. Dr. Schyve pointed out that consensus on minimum pieces of information inevitably will change as the science behind it changes.
Dr. Chin discussed Kaiser-Permanente (K-P) in terms of the business case for standards. Because of the integrated nature of health care and of the electronic Results Reporting system at K-P, Dr. Chin stated that the systems of that group are de facto standards. Unique patient identifiers make it possible to interface, although imperfectly, with the transcription systems of outside providers. This information system is comprehensive and "patient-centric," and allows for a variety of functional and subject-matter views of the data that permit individual case management as well as follow-up for patients due for certain tests.
Dr. Chin described the three components of K-P's CPR: Results Reporting, prevention and disease registries, and Epicare. Epicare is the comprehensive, computer-based outpatient record, which also both computerizes all the processes that occur in the outpatient arena and integrates decision support. Health care is guided by the medical group, not an insurance company.
The de facto standards at K-P include a single patient identifier, single systems, and complete information on members. These enable consolidation of information, which results in improved quality and appropriateness of care, reduced unnecessary variation, and reduced cost.
Dr. Chin presented K-P's view of appropriate priorities for health care data standards: unique patient identifier; dates, labels, and bodies of information; test and procedure naming standards visible in summary form; results standards for comparison of results between systems; and diagnosis codes. It is not necessary to code everything, but only those elements with known benefit in both decision support and outcomes analysis.
The technology XML allows one to structure and transfer a chart from one system to another at various levels and degrees of specificity.
Dr. Mattison discussed a number of outstanding problems and issues to be addressed from the perspective of an integrated delivery provider: (1) overarching architecture into which simpler standards would fit, and (2) coordination of disparate standardization initiatives. Strategic objectives would include: (1) create a unified, longitudinal record to support coordinated care delivery and work flow between institutions over time; (2) create clinical data as substrate for advanced decision support; (3) create clinical data as substrate for population- based outcomes analysis; and (4) combine objectives (2) and (3) into support for continuous quality-improvement cycles. Dr. Mattison described three prerequisites for fulfilling the four strategic goals: (1) user interaction model; (2) terminology model with unique terms, based on business needs; and (3) information model, based on an industry standard information model such as HL7, supporting transfer of full semantic richness of notes across applications, platforms, and institutions over time.
Dr. Mattison then described his view of major shifts in the future: (1) Coding will shift from coders to clinicians; (2) responsibility for care will shift from a provider-centric model to a consumer-centric model; and (3) the universe of knowledge will shift.
Business incentives for health care providers, software developers, payers, employers, consumers, and others, exist in the marketplace to achieve these goals. Convergent incentives include low-cost, high-quality health care, and short-term return on investment. Divergent incentives relate to (1) commercial product differentiation versus adoption of standards, (2) time to market versus high-quality health care, and (3) rapid return on investment versus size and cost of the task.
Other than assurances of protection of privacy and confidentiality, major outstanding issues include: (1) mobility of consumers requiring representation of information across institutions; (2) annual 10-20 percent patient turnover in jobs and therefore health care providers; (3) enormous change management requirement to implement new systems, and expectable resistance (suggested regulatory prescription: set time tables; but significant direct consequences on personnel work flow will serve as an impediment); and (4) gap between the magnitude of the task and the need for incremental demonstration of return on investment.
Dr. Mattison suggested opportunities for regulatory influence: (1) identify conflicts and gaps in existing standards initiatives; (2) identify unleveraged opportunities; (3) endorse or fund specific solutions after moderate consensus is achieved, particularly in terminology and with an eye to the transitional problems encountered in translating from various European currencies to the new Euro and back; (4) address through regulatory activity market incentives to accelerate adoption and implementation of standards; and (5) engage consumer advocacy groups in the work on privacy and confidentiality issues.
Ms. Narcisi referred to an AMA handout that defines the association's policy on the CPR. She asserted that no federal legislation requires implementation of medical record standards, and it is unclear whether HIPAA applies to a single, large system. The AMA recommends that: (1) NCVHS continue to study and identify privacy, security, and access measures to protect individually identifiable health information; (2) physician input be emphasized in the decisions on standards; (3) standards be developed in the categories of authentication, system interfaces, and interoperability; (4) existing incompatibility among systems be addressed, particularly in a small physician's office; (5) a true private-sector approach to standards development be undertaken, with the government role limited to such activities as helping to identify gaps in standards and facilitating and setting priorities for standards development; (6) the cost of implementation in terms of both dollars and productivity be recognized; and (7) federally mandated standards be avoided as an inappropriate and unproductive approach.
Ms. Narcisi asserted that standards will evolve in the private sector and automation of clinical data will occur at different rates throughout the health care industry. She noted that needs of physicians and patients may be different, and that the record systems be continually evaluated. She urged accelerated progress in the vendor community focused on interfaces, interoperability, and authenticity.
The AMA believes that CPRs are deficient in a number of areas, such as medical vocabulary (a possible solution is interlocking code sets), identifiers, data exchange, and privacy and confidentiality. AMA is currently working on the CPT5 project, the work of which Ms. Narcisi explained. The AMA believes that data content standards for claims encounters and attachments should adhere to the recommendations of the National Uniform Claim Committee, the National Uniform Billing Committee, the National Council on Prescription Drug Programs, and the American Dental Association.
Ms. Narcisi stated that although the four focus areas identified by NCVHS should be studied, in addition to studying and identifying privacy, security, and access measures.
Dr. Orient referred Work Group members to her written testimony. She proposed an alternate way to look at the issue of CPR: perhaps there should not be a nationwide standard for the electronic medical record. She asserted that many independent physicians do not wish to be bound by its constraints, that patients do not want to be faced with the additional charges, and that patients do not trust that their private information would remain private.
She cited the difficulties in accomplishing such a task, pointing as examples to the efforts of mathematicians who for centuries have been unable to accomplish "squaring the circle" because of a "little irrational number . . . called pi," and to the Harvard Law of Animal Behavior, where you "put an experimental animal under carefully controlled conditions, and the critter does whatever it pleases." She asserted that those involved in this effort have assumed that (1) quality is the same as compliance with a set of criteria and standards; (2) we know the best way to take care of every clinical problem; (3) "when we are measuring quality, we really are measuring something [when] actually, we're measuring what we can measure"; and (4) "we need to have a top-down command and control economy in the medical system" and a "top-down rationing system that will allocate resources."
Dr. Orient suggested that the role of government should be to protect individual rights against their violation by others, and not to dictate the standards of medical care or to dictate how medical records shall be kept. This approach, she asserted, destroys the patient-physician relationship, impedes scientific progress by trying to put everyone in a straight jacket, will drive independent practitioners out of the medical field because of overhead costs, and will seriously degrade the quality of medical care.
Dr. Spackman introduced the College of American Pathologists' (CAP) initiative SNOMED, Systematized Nomenclature of Medicine, a comprehensive, controlled reference terminology designed to encompass all terms used in medicine in order to ensure delivery of quality clinical services. SNOMED-RT, to be premiered in June 1999, will be available as a foundation for the CPR. Dr. Spackman asserted that SNOMED can contribute to improvement in patient care, reduction of errors inherent in data coding, facilitation of research, and support of compatibility across software applications.
In response to the Work Group's prepared questions, Dr. Spackman stated the following: (1) Congress appears to recognize and support the private sector's leadership in electronic exchange of health care information. (2) Factors that may delay development and widespread implementation of uniform standards include lack of understanding of different types of terminology and problems of scale. (3) SNOMED has worked cooperatively with a number of organizations in the private sector to develop, maintain, and coordinate terminologies; the financial ability to continue to do so is uncertain. Without government support, widespread development and implementation of uniform standard terminologies will continue to be delayed. CAP recommends a public/private partnership in the development of ongoing maintenance, for which government involvement is crucial in testing and funding. (4) SNOMED, HL7, and DICOM would be important components in patient medical record information, as would ICD-9-CM and CPT4 and their successors. (5) CAP believes the four focus areas are appropriate; but coordination, maintenance, continuous availability, interoperability, accountability, and clinical specificity are key issues.
Dr. Spackman stated that international collaboration and convergence is an important factor, and CAP is engaged in such efforts.
Dr. Orient, asked for her recommendations to Congress, responded that she would point out the potential adverse consequences of mandating nationwide a deeply, fundamentally flawed system; some may find such systems useful, but use of the systems should not be mandated.
In order to improve the situation, Dr. Orient asserted that "truly voluntary" participation should be possible; payment for services, for example, should be possible in alternative ways. Dr. McDonald stated that the CPR will not be forced on anyone.
Dr. Mattison described the pattern of change resistance, which typically diminishes within months of the introduction of a new system.
Ms. Narcisi stated that the AMA opposes mandating standards, which should evolve in the private sector, perhaps according to priority guidelines from the government. Information would be helpful, for example, on how coordination is proceeding between the Departments of Defense and Veterans Affairs in those kinds of projects.
Mr. Mattison described the benefits of using HL7 as a standard: All providers of health care to a consumer would have the benefit of all information that consumer had shared with the health care delivery system, which would result in a well-informed decision on care. Vendors could focus on a single standard that would facilitate support of multiple different implementations in different institutions, resulting in lower cost. Normalization of data across different applications would be enabled, thus accelerating the ability to do population-based outcomes analysis. Faster quality improvement cycles would be enabled.
To Mr. Blair's question on the role of government in SNOMED, Dr. Spackman responded that government could help by (1) facilitating collaboration with a broader universe of developers and users of terminology, (2) funding analysis or testing of the tools and processes for convergence, (3) adopting standard clinical reference terminologies within government projects, such as the National Cancer Institute, and (4) studying and testing reliability and useability of the encoding process.
Mr. Mayes pointed out that the federal government itself is the largest single payer, provider, and researcher of health care in the United States. It is likely that the standards will be used in federal programmatic requirements, and that they will have far-reaching effects.
Ms. Narcisi stated that the AMA is convening advisors to consider the optimal content of a medical record and then to develop guidelines to help physicians select systems.
Dr. Orient asserted that building a national infrastructure for CPRs in the private sector relates to an initiative involving immunization records. She suggested that a long-term pilot project related to CPRs be carried on by unbiased observers to see the possible effects.
Dr. McDonald asked the panelists to comment on reducing barriers to widespread adoption of CPRs by reducing the cost of standards products to users. Dr. Orient cited the CPT code, which is only available for sale and is neither downloadable from the Internet nor available from the Superintendent of Public Documents, as a great expense for physicians. Dr. Spackman stated that helping to fund distribution of documents is a "good thing," but that the real cost in clinical terminology has to do with the integration and the scale. Ms. Narcisi stated that many groups have difficulties dealing with this issue, but that costs are likely to diminish if vendors can be motivated to incorporate the standards into the system; interoperability will make a difference. Dr. Mattison cautioned that government intervention in standards could occur should the government become involved in subsidization of standards initiatives.
Dr. Gardner referred to a set of handouts and stated that the opinions he expressed would be his own. Dr. Gardner described the information systems in his hospital, which are characterized by choice of paper and/or electronics, and by a computer terminal at every bedside. Internal studies have shown the value of such computerized systems as computer- assisted antibiotic management, which has resulted in a decrease by about 23 percent of antibiotic use, a decrease in the mortality rate, a decrease in adverse drug events, and a reduction in the cost of antibiotics of about 42 percent.
He noted that problems still exist in terms of information quality, and that the use of highly computerized systems does not necessarily solve these issues. He also stated that the use of a GPS system in the trucking industry is more advanced technologically than his hospital's patient locator capability. He pointed out how the delay in standards development affects his hospital's operations: For some systems, nurses read computer data from one piece of equipment and hand enter it into another computer.
Dr. Gardner stated the need for a Board of Standards to ensure that changes in standards are made appropriately as science and medicine change over time.
Mr. Arges stated that although improvements in computer processing, storage, and communication capabilities over the past several years have occurred, with a corresponding decrease in the cost for acquiring the enabling technology, providers generally find it difficult to obtain the financing to undertake a massive information system investment.
In response to the prepared questions, Mr. Arges stated the following:
Mr. Bozeman described his organization and its CPR efforts over the past 15 years. Patients have common patient identifiers and use magnetic stripe patient identification cards. All charting, except by a few physicians, is done by computer.
In response to the prepared questions, Mr. Bozeman stated the following:
In response to questions from Dr. Cohn, Mr. Bozeman reiterated that although his organization is committed to the use of electronic information systems, many are not. He also described his organization's system of databases, one of text (primarily physician's notes) and one that is the Cost Information Decision Support System.
To Dr. Lumpkin's question on the role of incentives in the marketplace, Dr. Gardner responded that incentives to conduct demonstration projects would be valuable. He also stated that decisions on codes for drugs and lab tests would be useful.
Mr. Arges concurred with Mr. Bozeman that federal funding cuts are impediments to moving forward on standards development. Mr. Arges stated that no formal study is underway by the AHA on this issue, although some members are working independently on the CPR. Providing resource materials would be valuable in helping members understand how CPRs are being applied by others.
Mr. Bozeman stated that the patient common identifier in his organization is not the Social Security number, but that the Social Security number is part of the data collected on patients and can be cross-referenced in an attempt to identify a patient who does not present an ID card.
In response to a request for comments from the public, Mr. Henry Hefferin, suggested that the Committee look into funding the National Committee on Information Technology Standards T2 committee's work on consensus semantics and artificial language. He stated that he will provide more information to the Committee.
Mr. Arges asserted that hospitals want more standardization in records, and systems that would be turnkey operations. He stated that hospitals would object if the government mandated using the standards ultimately developed, because changes in technology might result in additional costs. When organizations merge, Mr. Arges continued, inconsistency in data standards pose problems that could be ameliorated by implementation of HIPAA standards; however, most such merging organizations are currently struggling with the business issues of integrating the new entity as part of their operations. Mr. Arges stated that although many hospitals submit flat file specifications, to move to a different format would be costly with no benefit.
In a dialogue with Dr. Lumpkin, Mr. Arges asserted that government mandating of a standard format for electronic transmission from, for example, a hospital to a doctor, would be palatable, as distinct from government mandating how a hospital should run its CPR or medical record repository. Communication using standard formats between hospitals and physicians who may be "captive" may be an important issue, as would be legal concerns related to reliability of information provided.
Asked by Dr. Fitzmaurice how LDS Hospital gains the cooperation of physicians with the computerized system in use there, Dr. Gardner explained that demonstrated time savings in accessing information is a significant incentive. A recent survey among LDS Hospital doctors showed no great fear about "Big Brother." Automating the manual system whereby a physician scribbles a little electronic note and someone else expands on it is a critical element of user friendliness.
The meeting was recessed until the following day.
Dr. Cohn reconvened the hearings.
Dr. Russler referred to his written statement and responded to the Work Group's prepared questions. The HBOC interpretation of Congress' instructions is that the mandate on health care standards is analogous to the recent FCC process to create high-definition TV (HDTV) standards for the broadcast industry. Dr. Russler asserted that only when political forces determine which standards are to be endorsed is the business case adequate for investment in communication of guidelines and patient record information by vendors of information systems.
Factors that produce delay in implementation include: (1) Lack of investment by government and other stakeholders in standards that specifically incorporate cost and quality data from evidence-based medicine into the useful vocabulary choices in information models. (2) Redirection of current vocabulary investments into nonvocabulary administrative functions. Regulation and audit of voluntary standards organizations are necessary to protect investments in development of the vocabularies. (3) Legal copyright restrictions. Use of proprietary vocabulary may limit the ability of a health care organization to receive patient care information from other organizations in emergencies.
The government should participate in the development of standards on health care data communication to the same degree as the FCC has done in the HDTV process. Government support is appropriate to facilitate consensus building on vocabularies. HBOC supports the developing partnership between HL7 and X12 for communication of claims attachments between payors and providers.
Dr. Russler suggested that combining focus areas 1 and 2 may improve efficiency of analysis. A second alternative would be to complete area 2 before beginning area 1. Still another would be to combine 1 and 2, but to identify focus areas in facilities location, health care participant, and clinical and administrative event modeling. Vocabulary standardization should not be separated from information model standardization. Focus areas 3 and 4 are well identified.
Dr. Tang responded to the Work Group's prepared questions. HIPAA legislation asks NCVHS to recommend how government can accelerate the development and use of CPR to apply to patient care, research, and public health. Dr. Tang's views of government involvement include setting the bar aligning the incentives and jump-starting the process: (1) requiring reporting on quality using standardized measures and a fixed time line imposed by either legislation or reimbursement policy; (2) establishing incentives for providers to gather and report on their compliance with quality standards and to share the cost of implementing and maintaining CPR systems; (3) establishing and funding projects to accomplish the work, such as the GCPR Project.
In answer to the Work Group's prepared questions, Mr. Tonks stated that his company's interpretation of the congressional directive is to focus narrowly on the need for data standards required in the patient medical record and its electronic exchange, although other areas of focus should have higher priority and could probably provide greater return on investment. He expressed frustration with the lag in information exchange behind that in certain other industries.
An important factor inhibiting development of uniform standards is a lack of understanding of the data term or code placed in each data field of a message, which impedes plug-and-play interoperability. Mr. Tonks stated that support by the government would be valuable in this area. Many organizations must work together to resolve interoperability issues, and the government must take a leadership role in the following ways: (1) define and mandate standards needed for government programs such as DRGs, CPTs, and ICDs, to include episodic groupers and other standards, all in the public domain, that will aid in the management of patient care over the lifetime; (2) contribute input to industry standard organizations, to be subjected to the consensus process with other inputs to achieve a common standard; (3) fund research efforts in support of standards organizations and provide a forum for collaborative efforts among those organizations; (4) implement industry-accepted standards and mandate them in all government projects and contracts, and at least encourage them in the public sector elsewhere. Mr. Tonks also defined his view of the roles of standards and lexical content organizations, and of vendors.
Mr. Tonks asserted that for exchange of electronic messages, the government should provide input to the definition of the reference information model to ensure that the model meets its needs, and it also should be more active in defining a reference terminology model. To be compliant, any system must translate only its internal information and terminology models to the standard to be able to send or receive messages to or from any other compliant system.
In a discussion of the four focus areas, Mr. Tonks suggested that the government fund establishment of a terminology service bureau to create a library or reference terminology model of medical concepts, each containing an identifying number mapped to all HL7- approved content, coding, and nomenclature provider terms in code sets such as SNOMED and LOIN. Updates from the model and library would be available by subscription. The bureau would represent a single standard lexical source requiring developers to sign only one contract that would provide plug-and-play content for all HL7 messages. Mr. Tonk suggested that an ideal pilot for creation and testing of the bureau would be the GCPR Project.
Dr. Middleton referred to his written testimony. Dr. Middleton asserted that his company's growth is attributable largely to the use of standards in creating content in software usable by a wide variety of customers. He asserted that a CPR must support or enhance the doctor-patient relationship; improve delivery efficiency, quality, and service at appropriate cost; empower clinicians with state-of-the-art tools to support clinical information access in decision making; provide the ability to exchange longitudinal, complex records; and assure secure access and confidentiality of data.
Dr. Middleton presented a macro-economic cost/benefit analysis and data on opportunities for cost savings with health information technology (IT), both qualitative and quantitative. He also discussed return on investment.
In response to the Work Group's prepared questions, he noted that HIPAA's mandate is to adopt national standards as necessary for efficient electronic, administrative, and financial health care transactions, to require certain public-sector-oriented entities to implement standards, to provide privacy legislation, and to provide recommendations for a CPR. The mandate also applies to standards for EDI administrative transactions. Incentives for diverse stakeholders in the industry are not necessarily congruent, and this misalignment makes decision making concerning the customer base difficult for vendors. He also asserted that the standard-setting process is grossly uncoordinated, which also poses difficulties for vendors; that vendors poorly implement and coordinate the implementation of systems; and that the laudable efforts of professional societies to standardize are nevertheless uncoordinated.
Dr. Middleton proposed three counter-questions to the third prepared question: Why can't vendors do it themselves? Why can't providers do it themselves? Should the government assume an expanded leadership role in the field? He asserted that it was possible for the U.S. national standard-setting efforts to establish standards "in a reasonable way" for standards that are quantifiable in nature, but it is more difficult or impossible to do so for soft standards (e.g., terminology, modeling). The government role relates to the size of the problem, he noted, and the leadership void first identified in the 1993 GAO report remains unfilled.
Dr. Middleton suggested simplification of the goals stipulated in the fourth question to the bare minimum with respect to identifiers, terminology, data sets and models, interoperability, connectivity, security, and confidentiality. He also suggested establishment of a framework for comprehensive health terminology systems and an agency to act as a responsible coordinating party. Dr. Middleton identified critical issues for standards in informatics, including the following: (1) an educational and training problem related to computing; (2) establishment of documentation standards for paper and electronic records; and (3) development of minimal data sets. He asserted that overcoming in the short term the hurdle of interoperability standards should be attempted, and that the goal should be to develop a national health care information infrastructure that is a secure national network for communication of electronic information pertaining to health, well being, reimbursement analysis, etc., while protecting the security and confidentiality of individual patient identifiable information.
Dr. Middleton seconded Dr. Tang's proposal to mandate quality standards, stating that such a mandate would immediately create demand for compliant systems. Dr. Middleton also suggested mandating the assessment of drug interactions at point of care, thus moving knowledge bases from the pharmacist and downstream assessment of drug interaction upstream to the point of care. He suggested as the most important standards those related to documentation, interoperability, outcomes reporting, and clinical-decision support terminology representation.
He asserted that the business case is clear for those organizations at risk for their health care, noting the need for further pilot studies to address longer-term quality effects of these kinds of systems on the health and well being of the nation at large. He concurred with the four focus areas, and seconded Mr. Tonks's idea of a terminology service bureau maintained by a trusted agency.
Panelists cited data that supported a range of cost savings as a result of electronic transfer of patient records. They next discussed steps the government could take to encourage better alignment of incentives. Dr. Tang suggested mandating quality-related groups, which would define the need for data, which in turn would define the need for data standards, and that would align decision makers. He cautioned against the broader continuum of care information systems reverting to the billing world.
Dr. Middleton stated that although Dr. Tang's suggested task may be beyond the scope of the NCVHS Committee, some marketplace activities are positive: (1) The trend from fee for service to managed care is continuing toward consumerism, personal choice, and accountability, and (2) quality data are available to consumers for decision making. Dr. Russler discussed his view of CPT coding as quality standards, and disincentives of investing in standards-based development until the standard is established. Mr. Tonks suggested that DRGs would be useful in quality and delivery of care.
Mr. Mayes stated that HCFA is moving into the area of quality care information. He asked the panelists their opinions of HCFA's idea to shift from a paper-based to electronic specification for their requirements, an object type of specification whereby they pre-interpret what is wanted, "wrap some APIs around it," and make it available freely to the industry. He also asked whether the vendors saw any impact on differences among state laws.
Dr. Tang stated that vendors are affected by state laws, and he termed HCFA's proposal an outstanding idea. He advised HCFA to be specific and precise with definitions of data elements and the methods by which they are gathered. Dr. Middleton seconded Dr. Tang's comments and urged the government to move ahead quickly on a quality reporting plug-in. Dr. Middleton questioned whether in the future, medicine will be state regulated or if it would be worthwhile to consider federal models such as the VA has used; federal superseding legislation would be helpful. Dr. Russler stated that his organization must abide by state regulations as well as international, and stated his hope that standards also would be adopted internationally. He noted that the more uniformity present, the lower the cost to deliver products to the customer. He encouraged coordination of the template proposal in HL7. Mr. Tonks asserted that better coordination among the stakeholders and more leveling and definition of ownership of data that can be shared will be valuable. Dr. Cohn requested that the panelists enlist their legal departments to send to the Work Group additional information on issues related to state regulation.
To Ms. Fyffe's question concerning incentives for average hospitals to shift to CPRs, Dr. Tang suggested making this shift a business imperative. It is better to have quality, rather than coding, drive the business imperative, he asserted. Dr. Middleton suggested that attention to Y2K issues is consuming many businesses at present, that incentives would help to move the industry forward, and that some segments of the marketplace are already clamoring for this type of support. Dr. Russler's experience is that many hospital administrators are interested in electronic systems, but probably not those entities with institutional financial problems. In home health care, 15,000 to 20,000 nurses carry laptops into homes, and demand for such systems is high. One problem is that these systems cannot communicate with each other. Mr. Tonks pointed out the significant savings that can be realized by avoiding retaking of lab tests, etc. Dr. Lumpkin expressed interest in visiting some successful hospital installations.
Dr. Fitzmaurice described the efforts of the Agency for Health Care Policy and Research (AHCPR) and the voluntary ANSI Health Care Informatics Standards Board in the arena of health care quality research and in coordinating of standards, respectively. Mr. Tonks believes the effort needs more than voluntary coordination. One need is for coordinated public relations; another is for a requirement that the participants attend facilitated meetings that would coordinate specific domains and responsibility for them. The marketplace should be free to implement any standard, but the standards should be consistent across all organizations.
Dr. Middleton described the real costs, significant particularly to small and mid-sized companies, of sending volunteers to meetings in remote cities, particularly given the slow pace of the process. He asserted that his company cannot wait for standards to be formulated and is proceeding to implement "robust-enough" standards; they must "do what we can to survive today and meet the needs of the customer today, and we participate in those organizations from afar sometimes." Dr. Middleton suggested that more government guidance and creation of a sense of urgency might be necessary in the "soft standards" development process. Dr. Russler referred the Work Group to the analogy he drew to the HDTV standards process, in which "somebody had to make a decision"; parts are voluntary, but at the end, the FCC said, "This is the standard." Both carrots and sticks are necessary. Dr. Russler suggested that required participation in standard setting would level the playing field. Dr. Middleton also supported equal financial aid to level the field. In order to accelerate the process, Dr. Tang suggested imposing a deadline. Dr. Fitzmaurice added that since 1991, AHCPR has funded meetings of HISB and partners with the other standards developing organizations in funding meetings in different parts of the country.
As a way to avoid pitfalls in the process, Dr. Russler suggested looking at successful standard-setting programs of the past and use those processes to justify actions that this group takes because it is best practice.
Dr. Ferrans asked whether the development of standards will accelerate development of software, which would result in downward price pressure on CPRs. Dr. Tang replied that imposing the reporting of quality metrics would turn that activity into a business imperative for decision makers to decide to enter the process of implementing change, to capture clinical data in the process of care. Vendors need help in motivating buyers to involve themselves in the process of change. Dr. Middleton seconded Dr. Tang's response on implementation. Dr. Middleton stated his preference for having standard tools with which to build a system as a customer requests it; he would prefer not to be told how to code the system, which software language to use, or how to implement the human-computer interface, etc. Dr. Russler's company can demonstrate that the current costs of their products give a favorable return on investment, which helps customers to buy it. Without standards, he noted, the cost of creation of the product is much higher, the customer would need to pay higher fees, and health care suffers.
Mr. Tonks asserted that government assistance is called for in developing reference models and reference libraries. The implementation of the model(s) should be conducted by the free marketplace.
All panelists supported Ms. Fyffe's idea that medical schools must offer a required course dealing with standards of documentation of medical records; Dr. Russler suggested that the course also should be required of today's chief medical officers of health care organizations.
The panelists discussed the technical aspects of reference terminology, described their support for SNOMED-RT and other efforts, and identified certain limitations in reference terminology.
Dr. Lumpkin asked the panelists for clarification of the seeming paradox that health care providers do not want regulations and vendors do. Dr. Tang suggested that two different levels are being addressed. Standards for quality would motivate and mobilize both the buyers and users of the systems, and the people who write standards to make accelerated progress. Panelists were asked to elaborate on their views on this issue in writing.
Dr. Morris referred to his written handout and addressed the Work Group's prepared questions. He asserted that although it is clear that Congress wishes to adopt uniform data standards for patient medical record information, the parts of that information to be automated are not clear. Efficiencies are to be gained by capturing data at the point of care, a model that has proven itself in such industries as telecommunications and banking, but it is unwise to force clinicians to use data standards, such as those for messaging and transactions, that should only be used internally in the software process. Clinicians should only need to use content- relevant standards.
Factors that delay implementation of standards include the following: (1) dissociation of the content and messaging formats that we use, and examining which standards focus on which areas and how that is done; (2) insufficient focus on the customer/user; (3) lack of consensus on what is a CPR and what should be the components of a chart; and (4) transition from human-readable processes to machine-readable processes.
Dr. Morris described the type of guidelines that should be supported: centralized code- assigning authority, in the public domain at lowest cost, participation in code-assigning process by all significant stakeholders, frequent and continuous updates (forward and backward compatible), series of user-friendly tools for determining coding schemas, nondeletion of identifiers, and attention to appropriate domains. He suggested as a model the Human Genome Project of government, industry, and academic collaboration. In this project, centers of excellence and people are focused on different areas, but they agree on a common model and drive toward a common synthesis.
Dr. Morris stated that the Committee has the opportunity to set the framework for the process and the infrastructure. An industry-specific partnership with government and academia should evolve and migrate the project.
Dr. Peters addressed the prepared questions. He stated that he interprets the congressional instructions to be advisory (rather than compulsory) in nature, with the objective to foster exchange of information of patient and provider benefit in the best way to obtain the benefits of standardization.
Among the factors preventing implementation of standards are the following: (1) inability of the health care information systems industry to produce software valuable to physicians and health care institutions, (2) resistance from many health care providers, (3) issues surrounding how payors pay for health care, (4) trend toward outpatient care, (5) trend toward basing medical investment decisions on the current cost of providing care, and (6) extremely expensive infrastructure.
Dr. Peters cautioned against both too strong a private-sector role in the standards businessÄdue to proprietary concerns and his perception that some individuals are trying to make careers of the standards processÄand repetition of the problems encountered with the CHCS project. Dr. Peters suggested that the government mandate the private sector to complete the standards work, noting that private-sector involvement beyond the health care IT segment may be necessary.
Essential standards include interoperability, and discussions on common data sets must include the medical subspecialty societies. Dr. Peters cited his own and others' support for XML SGML as a standard for uniform syntax. Uniform document content and definition of health care documents and their structure are also under consideration, and measures are in place to coordinate the efforts of various discussion groups. For terminology and coding, Dr. Peters stated that although he supports the SNOMED-RT effort, it does not seem practical for vendors, who need "really practical terminology, things that they can use . . . to exchange content." He suggested that a focus on data and content exchange will be more valuable than a focus on modeling and data models.
Dr. Peters supported the business case for rapid standards development by stating that physicians, who constitute the majority of decision makers in the market, are receptive. He expressed his agreement with the four areas of emphasis, reiterated his support for a proactive government role in standards setting, and urged the Work Group to focus on content and move away from consideration of code sets.
Dr. McNamara stated that many organizations in the health care IT industry are challenged not so much by the absence of standards in medical vocabulary as by the "proliferation of numerous disparate, incomplete pseudostandards" that lack either the clinical depth or wide acceptance in the industry needed to facilitate interoperability.
He focused his presentation on drug vocabularies, an area in which he stated that significant rapid achievements may be realized with modest effort, resources, and leadership from the government. The May 7 NPRM proposed that national drug codes (NDCs) become the national standard for transactions requiring drug codes. Dr. McNamara referred to his company's in-depth comments on the NPRM, which included a list of 13 proposed criteria for a drug vocabulary system and specific recommendations for action and regulatory language (available on the DHHS Administrative Simplification website).
In summary, the NDC system is hindered by a lack of central code assigned authority, lack of a common interorganization meaning for each NDC, and absence of NDCs for a large number of drugs, among other deficiencies. Importantly, because the NDC system is excessively granularÄthe assignment of codes occurs at the level of the drug packageÄthe system design is inappropriate for the needs of most providers who prescribe, dispense, or administer drugs, or for those who handle administrative functions involving data aggregation and analysis. Burdensome administrative activities would be required if NDCs were required for electronic transactions involving drugs.
Dr. McNamara listed the 13 proposed criteria for evaluation of code sets, including, for example, maintaining code sets by a central code-assigning authority, availability of the code set in the public domain at low or no cost, code assignment process open to review, and frequent, regular updates to the code set.
He described four possible approaches to implement a universally accepted drug vocabulary. Potentially the best approach, he asserted, is for government to work with industry in creating or selecting a robust drug vocabulary that addresses the business needs of organizations and individuals who wish to send drug-related transactions covered by HIPAA. Government might select a specific proprietary vocabulary and mandate its use, and then subsidize its continuing maintenance and/or charge a nominal fee for its use. End-user cost of nomenclatures should be an important consideration; Dr. McNamara offered the use of Multima Lexicon under a no-fee license agreement for developers and end-users.
Mr. Radtke described his company's involvement in all segments of health careÄas provider, health carrier, health care researcher, software vendor, and consumer of health care information. He stated that the government as payor can require the development of generic, flexible standards, common outcome measures, and common vocabulary.
In answer to the Work Group's prepared questions, Mr. Radtke observed that Congress is seeking advice on whether and how to use its mandate to provide for specific standardization in the industry. Ford believes that not all medical information need be standardized, but intervention is required in many places.
Among the issues delaying the development of standards is the variation in existing medical records, and the cost in rework to revise or replace software. Because of the weak incentive structure for collaboration, the private sector will not come together into consensus. Government's role is to help facilitate a more rapid and orderly transition to a uniform standard.
Mr. Radtke suggested that standards for medical records should focus jointly on data storage, the medical record itself, and data transmission. The four focus areas are important to Ford, but confidentiality and security of information are high priorities. Mr. Radtke asserted that a flexible data model with common definitions would be a good start; an inflexible standard that takes years to modify is a "disaster waiting to happen."
Dr. Cohn asked the panelists to advise the Work Group on steps the government should take to foster the alignment of incentives to develop a uniform standard. Dr. Morris suggested that major employers be encouraged to adopt a standard; a strong incentive would be for others not to be left behind. Dr. Peters concurred, stating that standards that focus on data interchange would be less restrictive. Mr. Radkte pointed out that requirements that the government imposes on its own facilities have great influence over the activities of other entities. Dr. McNamara responded that vis-a-vis drugs, for which everyone wants to have a low-cost list, the government's alignment of the small group of drug vocabulary creators can be accomplished by insisting on certain standards in that arena.
Dr. Peters expanded on his suggestion that a focus on codes be minimized and agreed to Mr. Blair's request to compile a paper on the topic for the Work Group.
Mr. Radtke elaborated on the differences in IT structures among the automakers: Ford focuses on quality, General Motors on finances, and Daimler-Chrysler typically outsources resources. The companies, in their Three-Auto Initiative, convene monthly; all are involved in hospital profiling and plan profiling.
Dr. Peters explained his regret that economic incentives, particularly those related to administrative functions, determine the relative importance of buying and implementing CPRs.
Dr. Lumpkin inquired whether the RFP process is sufficiently mature to be a feasible approach. Mr. Radtke suggested additional decision making is necessary before the point is reached where an RFP would be valuable. Dr. Peters stated that it might be possible in some instances (e.g., drugs) for a select group of stakeholders to convene and actually devise a standard without the necessity of an RFP; Dr. McNamara concurred that specific domains, such as drugs, would be more amenable to an RFP approach. Dr. Morris cautioned that the process or infrastructure must be established first.
Mr. Radtke stated that in-house, nonconfidential data is well understood in his company, but the data that comes from outside sources is coded differently, and handling and coordinating it are difficult. He told the Work Group that he was unaware of any standards- setting organization that was examining occupational health. Dr. McNamara stated that the drug nomenclature community in HL7 has reached an impasse, which federal efforts may prove to dislodge. Dr. Peters concurred that the differences among the vocabularies held by various companies is minimal, and that there is no reason for them to be logjammed on the basis of proprietary concerns. He added that an industry related to data exchange, integration engines, has grown up within health care that does not need to existÄand would no longer exist with government intervention. Dr. Morris stated his belief that XML will be a major player in the future, and he urged the government to look specifically at messaging content, syntax, and XML for an open way of migrating.
Mr. Groen suggested that while many consider the clinician to be the end user, others consider the patient to be the end user; this dichotomy is important in the context of resolving issues of interoperability and common terminology. Dr. Peters responded that the largest, most competitive market segment is the patient information, data management, and data entry industry, which is using HTML with web browsers as a primary format; most organizations are involved in the patient data-entry side, with transmission of data to doctors via e-mail, for example. Dr. Morris suggested that electronic interaction between provider and patient may be an issue in the future, an issue considerably different from what has been done in the past. Dr. McNamara brought up concerns about confidentiality and security of information, and of reconciling differences in records from separate providers for one patient. Mr. Radtke pointed out that the electronic future is rapidly approaching, and designs must keep this trend in mind. Dr. Ferrans stated that the patient-entered record could use a plug-and-play architecture, whereby it was an interface to the CPR. The patient, under privacy and confidentiality legislation, could validate the accuracy of the record.
Future dates for hearings include March 29-30 and May 17-18, 1999. The Work Group identified potential areas upon which to focus discussion in the future.
Mr. Blair suggested the following:
Dr. Cohn suggested the following:
Mr. Mayes classified the focus areas in three categories:
He also suggested that the discussion of the appropriate level of government involvement is important to continue.
Dr. Lumpkin's suggestions included the following to facilitate describing that which is necessary to define:
Dr. Fitzmaurice noted the following:
Mr. Garvie suggested that the true barriers and status quo be identified, perhaps using site visits at community hospitals to do so.
Ms. Fyffe described two visual images that the development of data interchange standards ought to make possible:
Dr. Ferrans discussed the following:
Dr. Fitzmaurice described the phenomenon of shifting work priorities as automation increases productivity. Salary savings would not be a fair measure of productivity, because jobs are not necessarily eliminated; workers engage in higher priority activities. Dr. Mayes described the difficulties in demonstrating a positive return in a 5-year time frame on HIPAA transactions due to the lack of both a cost model and data. Mr. Blair stated that early results of cost benefits for clinical improvements from CPRs are beginning to be available.
Ms. Fyffe reiterated the necessity for vendors to conduct adequate research and development work that will lead to interoperability of the approximately 2,000 different practice management systems in use in the United States.
Dr. Yasnoff pointed out the "remarkable consensus" among most of the witnesses that government should do something, particularly to facilitate and speed up the existing standards processes to coordinate them. He stated that although there was no unanimity among those witnesses on which approach to take, the literature reveals a number of important success stories. The CPR enables better analysis of health care and its potential for cost savings. Dr. Yasnoff also noted that no witnesses described any potential public health benefits of CPRs, such as the Centers for Disease Control having information and being able to act on it in a timely manner, and to plan for and detect bioterrorism events.
Dr. Cohn discussed the lack of clarity concerning what the business case ought to be, in part because of definitional problems. Mr. Blair pointed out the value of this session of hearings in identifying and clarifying a number of issues, and noted the necessity for additional information gathering, for analysis of the information, and ultimately for making (reviewed) recommendations by August 2000. The first step in the process might be to come to an understanding of the spectrum of terms "clinical data repository" or "computer-based patient record" or "clinical information system." An understanding of the differences, Mr. Blair asserted, will enhance development of the business case. Mr. Mayes questioned whether this exercise would be too detailed vis-a-vis the focus of the Work Group. Mr. Blair responded that it should be possible to quantify the benefits of moving to, for example, a higher level of interoperability. Ms. Humphreys suggested that dialogue on the issue would be enhanced by a common understanding of terms. Dr. Yasnoff suggested that the consultant community, which deals with many vendors and other organizations, might offer the needed insight to set up a framework to understand these relationships.
Ms. Fyffe pointed out the four parts to the mission of the Work Group: studying (identifying) the issue, adoption of uniform data standards (must define term first), patient medical record information (must define), and electronic exchange of such information.
Dr. Ferrans suggested framing the discussion in terms of looking at individual business cases for some data standards that could be uniform, to determine the net effect (benefits and costs to providers) of their adoption, and to present several examples of ways it could be done.
Dr. Cohn suggested that an analysis be undertaken of the ways in which the witnesses answered the first prepared question. Dr. Fitzmaurice indicated the need to map how the various standards discussed relate to application functions that yield rates of return, which would be of interest to Congress, DHHS, and the rest of the community. Dr. Fitzmaurice also suggested the need for understanding each of the itemized issues discussed earlier in order to be able to determine the level of constructive involvement of the Work Group. He proposed that a high priority of the work plan ought to be identifying the issues on which to focus.
Dr. Cohn posed the need to continue asking the country more questions about how the government can be of help. Groups from which additional comment should be solicited include the standards development organizations and the Departments of Veterans Affairs and of Defense. Ms. Humphreys and Mr. Blair concurred. Mr. Blair added that it would be necessary to pose more issue-focused questions on panels of standards development organizations and of users and of inputters of the data.
Mr. Mayes pointed out that several government organizations currently provide incentives: GCPR, HCFA, CDC, FDA, and NLM.
Ms. Fyffe suggested the necessity of including testimony from chronic-disease consumer groups. Mr. Mayes noted that a community of consumers interested in end-stage renal disease is developing an information system to link providers, payors, regulators, and consumers nationwide. Ms. Humphreys pointed out that some consumers concerned about privacy and confidentiality believe that they should carry their own medical records and permit them to appear on computers only for the briefest period of time.
Dr. Ferrans suggested that panels of standards development organizations be grouped by subject matter and be asked to agree, for example, on an appropriate role for government.
Dr. Cohn pointed out the necessity to establish a framework for the four items proposed by Ms. Fyffe. In addition to synthesizing the answers to Question 1, he recommended looking at the definition of patient medical record information and how it differs (or does not) from the electronic medical record. Ms. Fyffe suggested the necessity of offline discussions to define the scope of the Work Group. Dr. Fitzmaurice expressed his preference to focus on issues and not on what is a CPR. Dr. Cohn amended his recommendation, stating that the issue should be what is patient medical record information and what parts of it are important. Dr. Lumpkin stressed that the importance of adding value to patient care with electronic records be included in the discussion.
Following a brief discussion of the tentative agenda for the February working session, the meeting was adjourned at 4:08 p.m.
I hereby certify that, to the best of my knowledge, the foregoing summary of minutes is accurate and complete.
/s/Jeffrey S. Blair February 2000
_______________________________________________________
Chair Date