MEDINFO 2001 V. Patel et al. (Eds) Amsterdam: IOS Press 2001 IMIA. All rights reserved The Nursing Terminology Summit: Collaboration for Progress Judy Ozbolt a, Ida Androwich b, Suzanne Bakken c, Patricia Button d, Nicholas Hardiker e, Charles Mead f, Judith Warren g, Christine Zingo h a Vanderbilt University, Nashville, Tennessee, USA b Loyola University of Chicago, Chicago, Illinois, USA c Columbia University, New York, New York, USA d IDX Systems Corporation, Seattle, Washington, USA e Manchester University, Manchester, UK f Simione Central Holdings, Inc., Atlanta, Georgia, USA g University of Nebraska, Omaha, Nebraska, USA h Kaiser Permanente, Pasadena, California, USA Abstract The Nursing Terminology Summit has used collaborative processes to bring about significant changes in the development of terminology standards for nursing. This paper draws on agendas, reports, notes, and other documents from the Summit, in addition to the authors' own experience as Organizer, Steering Committee, and participants, to provide a brief history of the Summit process. The analysis identifies factors that increased the risk of failure as well as factors that fostered success. The paper concludes with lessons learned that can be applied in other arenas to promote change in medical informatics. Keywords: Terminology; Standards; Nursing Introduction Since 1999 the Nursing Terminology Summit has played initiating and contributing roles in significantly changing the development of nursing terminology standards. A series of invitational conferences and intervening work activities, the Summit has achieved its objectives through collaborative processes that reach across disciplinary and national boundaries. Accomplishments of the Summit include: Shifting the focus of nursing terminology standards development in the United States from expressionbased sets of terms to concept-based reference terminologies. Shifting the locus of nursing terminology standards development in the United States from disciplinary isolation to the mainstream of healthcare terminology standards development. Collaborating in worldwide efforts to develop a reference terminology model for nursing and to integrate that model into comprehensive healthcare terminology models, in the context of the International Standards Organization (ISO). Contributing to the development of European terminology standards for nursing by providing invited comments and critiques and helping to integrate the European standards work with the ISO work. This paper provides a brief history of the processes through which these changes have been accomplished. The analysis includes factors that increased the risk of failure and factors that contributed to success. Finally, the paper proposes "lessons learned" that can be applied in other situations to promote progress in health informatics. Materials and Methods In developing this history, the authors used electronic and printed materials, published and unpublished, related to the summit conferences and the intervening work activities. In addition, they drew upon their experiences in organizing, leading, and participating in the Summit. This account of the Summit therefore presents the perspectives of the Organizer (Ozbolt) and the Steering Committee (other authors). Results Nursing Terminology Standards Circa 1998: Impetus for Change By 1998, the American Nurses Association had recognized 7 terminologies as meeting its criteria for representing nursing care [1-6]. Evaluations showed that each served reasonably well the purpose for which it had been intended, but no single terminology was adequate to represent the full range of nursing concepts [7]. Moreover, these terminologies differed in purpose, scope, organization, method, level of detail, and conceptual foundation. Because 236
of these differences, the terminologies could not readily be combined into a unified nursing language [8]. Developers and users of nursing information systems were frustrated by the lack of a standard language to represent clinical phenomena and to capture comparable patient data across sites and times. Meanwhile, analyses based on new understanding emerging from work outside nursing revealed that although the recognized nursing terminologies were useful for certain purposes, they lacked properties needed for a reference terminology that could serve as a unifying standard [7, 9-13]. In the United Kingdom, Hardiker had demonstrated how formal modeling techniques could illuminate efforts to develop nursing terminology standards [14]. In the USA, a group calling itself the "Loose Cannons" had begun to apply terminology modeling techniques to represent nursing concepts, but nursing terminology developers generally had little understanding of this work or its significance. Apart from Hardiker and the "Loose Cannons" few people had deep understanding of both nursing terminologies and methods of standards development. Competitiveness among nursing terminology developers and parallel play between nurse terminologists and other terminology standards developers combined to stymie progress. The Nursing Terminology Summit: Method of Change To resolve the problems of nursing terminology standards clearly required bringing together all those responsible for maintaining or changing relevant aspects of the status quo along with those who had the expertise to define and create the needed change. An invitational conference seemed a logical way to start. Having conceived the idea of an invitational Summit Conference to resolve central issues in developing nursing terminology standards, Ozbolt first contacted Hardiker and the members of the Loose Canons and asked for their advice and support in the role of a Steering Committee. Together, this core group identified the persons who should be invited to the Summit. Preparation for the 1999 Summit Conference To define and accept a goal for nursing terminology standards that went beyond the competing sets of terms would require the participation of leaders from a number of health informatics arenas. This included the nursing terminology developers and the leaders of organizations sponsoring terminologies; experts in terminology standards development, many of whom were not nurses; leaders of professional organizations such as the American Nurses Association and the American Medical Informatics Association; representatives of United States government agencies concerned with informatics and terminologies; members of health care enterprises who were responsible for the use of terminologies in information systems; and representatives of the health informatics industry who were concerned with the use of standard terminologies in applications. Because the conference would require difficult and stressful work as participants were challenged to explore unfamiliar ideas and to adopt new positions, the Steering Committee determined that it would be inappropriate to have an audience. They identified about 40 persons as potential working participants at the conference. As planning began in autumn 1998, the Organizer and the Steering Committee did not know whether these busy and responsible people would be willing to attend a conference to work together to resolve difficult and sometimes contentious issues. They prepared a brief statement of the purpose and format of the conference and contacted prospective attendees by telephone, fax, and email. Those contacted were asked to fax back a statement that they would attend such a conference if funding could be obtained. Happily, nearly all said that they would attend. With statements of intent in hand from this group of nationally and internationally recognized leaders, the Steering Committee began to seek financial support for the conference. Solicitations to government agencies, professional organizations, and health informatics firms brought positive responses. Each sponsor named one or two representatives who became full participants in the Summit. The grants, contracts, and contributions of the sponsors paid for meals and lodging in a conference center, a modest travel allowance, and, for the first Summit Conference, honoraria for non-nurse experts. Prior to the conference, the Organizer prepared a briefing book from materials submitted by the nursing terminology developers and the experts on language and standards. Distributed to participants shortly before the conference, the briefing book provided an introduction to topics to be discussed during the conference. Process of the 1999 Summit Conference The 1999 Summit Conference established the pattern of events followed at subsequent conferences: an opening reception on the first evening, a full day of plenary sessions to provide a base of common knowledge and a shared orientation to the work to be accomplished, a day of work in smaller groups, and a final half-day of reports from the small groups presented in plenary session. The conference closed with plans for ongoing work. Every evening participants relaxed together at a dinner hosted by the corporate sponsors. The plenary work day was devoted to presentations by nursing terminology developers and by experts on language and standards. By the end of the day, participants found that they needed time to reflect on the new ideas they had heard before they would be ready to determine what new work needed to be done. The following morning, therefore, the group met again in plenary session to summarize what they had learned and to determine directions. Bakken led the discussion, in which participants recognized the difference between interface terminologies and reference terminologies. They acknowledged that the nursing terminologies recognized by the ANA were all (with the exception of SNOMED International) interface terminologies. The work of the Summit, they agreed, would be to develop a model on which a reference terminology for 237
nursing could be constructed. Small groups then volunteered to work on modeling nursing diagnoses, nursing interventions, and nursing observations (appraisals of patient states that might constitute assessment data, therapeutic goals, or patient outcomes). In the final halfday, participants reported on what they had accomplished and what insights they had gained. They agreed to keep working on the models they were beginning to shape, to report their work in conferences and publications, and to meet again the following year. Consequences of the 1999 Summit Conference As a result of the Summit Conference, participants continued to collaborate and extend their work. The groups working on nursing diagnoses and interventions used conference calls, the Internet, and opportunities to meet at other conferences to continue developing and testing models to represent the essential concepts and their relationships. Bakken and others who had worked on nursing observations tested the structure of clinical LOINC as a potential model for representing nursing concepts in observations. Ultimately, the LOINC structure was modified slightly to accommodate the nursing terminology. Collaboration with Health Level 7 (HL7, an organization that sets standards for communicating electronic messages in health care) led to the registration of nursing terminologies in HL7. Indeed, the Summit was especially successful in promoting collaboration across disciplines and employment settings and across groups associated with one or another of the recognized nursing terminologies. At least as important, the Summit not only brought nursing terminologies into standards-developing organizations but also led to changes in standards to accommodate nursing. Other important collaborations crossed national boundaries. Within three weeks of the 1999 conference, a paper on the Summit work submitted for the 2000 Triennial Nursing Informatics Conference of the International Medical Informatics Association (IMIA) was reviewed by the chair of the IMIA Nursing Informatics Special Interest Group (NI-SIG). She contacted the Organizer of the Summit and asked for the collaboration of Summit participants in her own initiative to submit a proposal for a New Work Item to the International Standards Organization. Based on the work of the Summit, the purpose of the New Work Item would be the development of a reference terminology model for nursing. Summit participants gladly joined with others from around the world, communicating by Internet to draft the proposal and to guide it through the ISO process under the joint sponsorship of IMIA and the International Council of Nurses (ICN). Meanwhile, Technical Committee 251 (TC 251) of the European Standards Committee (Comité Européen de Normalisation, or CEN) had mandated the development of a "categorial system for nursing" that had many of the properties of a reference terminology model. Summit participants who were active in the CEN work invited the other Summit participants, via the Summit listserv, to join the CEN Nursing listserv and provide comments on the developing draft. Many did so. At the Fall 1999 Symposium of the American Medical Informatics Association, the Organizer of the Summit met with leaders of the CEN effort and invited them to participate in the 2000 Summit. They accepted the invitation. Preparation for the 2000 Summit Conference Most of those who had attended the 1999 Summit also accepted the invitation to the 2000 Summit. Additional international participants associated with the ISO work also agreed to come, as did non-nurse U.S. experts who were working with ISO on terminology standards. The sponsors again provided support and participation. The extensive work accomplished since the 1999 Summit actually made it more difficult to set the agenda. Whereas the first conference had focused on developing a shared goal and finding ways to collaborate, the objectives of the second conference were more diffuse. The small group work carried on since the last Summit competed for time on the agenda with the ISO work and the CEN work. The briefing book contained updates of all these efforts. Process of the 2000 Summit Conference Participants came to the 2000 Summit full of enthusiasm for the work they had been doing together over the past year. The difficulty was to give sufficient attention to each initiative while determining how best to use the time together. Participants spent some time clarifying the mission of the Summit, which they identified as contributing to the development of a reference terminology for nursing. Both the CEN work and the ISO work were directed toward a reference terminology, and the Summit offered an opportunity for discussion among the leaders and principal authors of each initiative and a group of experts, the other Summit participants. The presence of HL7 leaders of vocabulary and modeling efforts meant that discussions also covered the compatibility of an eventual reference model with the HL7 Reference Information Model (RIM) and even the possibility of using the RIM as the reference terminology model. This last idea did not find acceptance at this meeting, but participants agreed that the reference terminology model would eventually need to be harmonized with the RIM. Meanwhile, it was important that the ISO work and the CEN work be compatible, not competing, although the timelines of the two initiatives might preclude resolving differences before the due date of the ISO work item. Small group work at the 2000 conference was devoted to testing the CEN model and the RIM for representing concepts in ANA-recognized nursing terminologies. In the final half-day, small groups gave reports and made detailed plans to continue their collaboration. Consequences of the 2000 Summit Conference Small groups continued to work together to test the various models for representing the concepts and relationships of diagnoses, interventions, and observations. Most strikingly, the ISO New Work Item was withdrawn and revised to incorporate the CEN work and resubmitted under the Vienna Accord with a new timeline and due date. Summit 238
participants continued to provide input to the ISO work, in pursuit of a single international standard. Plans for 2001 and Beyond Ongoing collaboration will be necessary to establish a reference terminology model for nursing, populate it with a reference terminology, keep the reference terminology current and compatible with other standards such as the HL7 RIM, reconcile the nursing reference terminology with other reference terminologies and standards, and respond to issues arising as interface terminologies are mapped to the reference terminology. The Nursing Terminology Summit will continue to provide a forum to increase understanding, make decisions, and work together across terminologies, disciplines, employment settings, organizations, and national boundaries. Discussion The Nursing Terminology Summit initiated important changes in the goals and methods of developing nursing terminology standards in the United States. Before the Summit nursing terminology developers were working in competition with one another and in isolation from standards developing organizations. Since the beginning of the Summit, nursing terminology developers have been working with experts on language and standards to create a reference terminology model and eventually a reference terminology and to integrate nursing into the mainstream of standards development. The Nursing Terminology Summit has also contributed to international efforts to develop standards for nursing terminology. By helping to shape the ISO work, commenting on the CEN work, and providing a forum for leaders of both to communicate with input from other experts, the Summit has helped to make these international efforts not disparate and competitive but collaborative and compatible. Many factors might have prevented this success. Participants might have declined to work together. They might have accorded too little importance to resolving the problems of nursing terminology standards to make the time and effort to come to the conferences. Financial support might not have been forthcoming. The agendas of the conferences might not have focused on the most urgent issues. Group dynamics at the conferences might have prevented fruitful work. Participants might not have made the time to work on terminology issues between conferences. Any of these factors would have made it difficult to change the status quo of nursing terminology development as it existed in 1998. Yet none of them did. Why has the Summit succeeded? First, the Organizer was willing to take a chance that the difficulties could be overcome. Second, long-standing, cordial, professional relationships established through organizations and conferences meant that the Steering Committee and the invitees were willing to take a chance too. Third, the leadership positions of invitees gave the sponsors confidence that the Summit could succeed in its mission and indeed enabled important decisions to be made. Fourth, the prestige of the invitees and the sponsors gave those who were less well-acquainted with the others confidence that their time would not be wasted. Fifth, the Steering Committee's expertise in nursing terminology standards development helped to make the agenda substantive and well targeted. Sixth, the Organizer and the Steering Committee paid special attention to group interactions and the exercise of diplomacy and tact. Finally, participants shared a commitment to developing the kind of terminology standards that will give meaning and validity to clinical data and enable their use to improve health care. Conclusions Some of the Summit's success factors can be applied elsewhere to solve problems and promote changes in health informatics. A few of the lessons learned are offered here: Bring together all those who are responsible for maintaining or changing the problem, along with those who have the expertise to define and promote a solution. Define the tasks and focus the work to maximize productivity while busy people are together. Relieve intense work periods with relaxing group events that build relationships. Keep discussions focused on ideas and facts, not on personal behavior or character. Even issues about which people disagree strongly can be discussed harmoniously if the discussions remain impersonal. Be willing to take the first step toward a solution and invite others to join the effort. Acknowledgements The Organizer, the Steering Committee, and the participants in Nursing Terminology Summit are grateful for the support of our sponsors. Agencies of the United States government that provided funding were the National Library of Medicine and the Division of Nursing of the Health Resources and Services Administration. The Nursing Informatics Working Group of the American Medical Informatics Association (AMIA) sponsored the Summit as a Working Group initiative. Sponsoring corporations have included Apelon, Inc.; Cerner Corporation; IDX Systems Corporation; 3M Health Information Systems; McKessonHBOC; Shared Medical Systems; and SNOMED International. References [1] Maas ML, Johnson M, Moorhead S. Classifying nursing-sensitive patient outcomes. Image: Journal of Nursing Scholarship, 1996; 28:295-301. [2] Martin KS, Scheet NJ. The Omaha System: Nursing diagnoses, interventions, and client outcomes. In N. M. Lang (Ed.), Nursing data systems: The emerging framework: Data system advances for clinical nursing practice, pp. 105-113. Washington, D.C.: American Nurses Publishing, 1995. 239
[3] McCloskey JC, Bulechek GM (Eds.). Iowa Intervention Project: Nursing Interventions Classification (NIC), 2 nd ed. St. Louis: Mosby, 1996. [4] Ozbolt JG. From minimum data to maximum impact: Using clinical data to strengthen patient care. Advanced Nursing Practice Quarterly, 1996; 1(4): 62-69. [5] Saba VK. Home Health Care Classifications (HHCCs): Nursing diagnoses and nursing interventions. In NM Lang (Ed.), Nursing data systems: The emerging Framework: Data system advances for clinical nursing practice, pp. 61-103. Washington DC: American Nurses Publishing, 1995. [6] Warren JJ, Hoskins LM. NANDA s Nursing Diagnosis Taxonomy: A nursing database. In NM Lang (Ed.), Nursing data systems: The emerging Framework: Data system advances for clinical nursing practice, pp. 49-59. [7] Henry SB, Warren JJ, Lange L, Button P. A review of major nursing vocabularies and the extent to which they have the characteristics required for implementation in computer-based systems. J Amer Med Informatics Assoc, 1998; 5:321-328 [8] Goosen WT, Epping PMM, Feuth T, Dassen TWN, Hasman A, van de Heuvel WJA. A comparison of nursing minimal data sets. Jour Amer Med Informatics Assoc, 1998; 5:152-163. [9] Cimino JJ. Controlled medical vocabulary construction: Methods from the Canon Group Jour Amer Med Informatics Assoc, 1994; 1:296-297. [10] Cimino JJ. Desiderata for controlled medical vocabularies in the twenty-first century. Methods of Information in Med, 1998; 37: 394-403. [11] Spackman KA, Campbell KE, Cote RA. SNOMED RT: A reference terminology for health care. In DR Masys (Ed.), JAMIA 1997 Proceedings Supplement: AMIA Annual Fall Symposium, 640-644. [12] Henry SB, Mead CN. Nursing classification systems: Necessary but not sufficient for representing What Nurses Do for inclusion in computer-based patient record systems. Jour Amer Med Informatics Assoc, 1997; 4: 222-232. [13] Mead CN, Henry SB. Documenting What Nurses Do Moving beyond coding and classification. In DR Masys (Ed.), JAMIA 1997 Proceedings Supplement: AMIA Annual Fall Symposium, 141-145. [14] Hardiker NR, Rector AL. Modeling nursing terminology using the GRAIL representation language. Jour Amer Med Informatics Assoc, 1998; 5:120-128. Address for Correspondence Judy Ozbolt, Ph.D., R.N., F.A.A.N., F.A.C.M.I. Independence Foundation Professor of Nursing and Professor of Biomedical Informatics Vanderbilt University Godchaux Hall 461 21 st Avenue South Nashville, TN 37240-0008 judy.ozbolt@mcmail.vanderbilt.edu 240