 |
The Data Dictionary Task Force (DDTF), chaired by Dr. Terri Monk of the University of Florida-Gainesville, and under the technical direction of Dr. Iain Sanderson of Duke University, announced the launch of data comparison software, Distributed Anesthesia Terms and Mapping System (DATAMS). DATAMS was developed through the collaborative efforts of the DDTF's two work groups, the Technical Working Group and the Clinical Working Group. DATAMS provides a mechanism for the Clinical Working Group to accept lists of data elements from multiple sources, including anesthesia information management system (AIMS) content and data sets or dictionaries developed by institutions or other organizations.
Drs. Monk and Sanderson presented the results of the DDTF's nine-month development process at the recent ASA meeting as well as meetings of the Society for Technology in Anesthesia (STA) and the Association of Anesthesia Clinical Directors (AACD). Since being charged by the APSF Executive Committee in October 2001 and generously funded by APSF, Deio, Draeger Medical, eko systems, GE Medical Systems, Philips Medical Systems, Picis, and Siemens Medical Systems in Spring 2002 the DDTF has made significant progress on its mission to help standardize data terms across anesthesia information systems. Along the way, the DDTF has won support from national and international professional anesthesia association and data standards organizations.
According to Dr. Monk, "The DDTF has been focused on delivering on its promise to provide a method by which data captured in commercial or homegrown AIMS could be compared and evaluated to improve patient care." Initially, the Clinical Working Group attempted to develop a flat list of mandatory data terms that could be imposed on anesthesia practice nationwide. The group was assisted by the submission of data elements from the DDTF corporate sponsors as well as input from Dr. David Reich and Dr. Homer Yang of the Canadian Anesthesia Society. "After several months of wrangling with spreadsheets and lists, we seriously considered locking ourselves in a room and 'divining' a master list. We quickly discarded that idea as impractical and unworkable when we realized that we could not convene a large enough group of experts who could abandon their practices for the amount of time needed to tackle the complexities of standardizing the data elements in the anesthesia record," explained Monk.
"The Technical Working Group drew on the experience of other groups that have labored over the same ground," noted Dr. Sanderson. "The DDTF gained valuable insight from ongoing efforts to create terminologies in other branches of medicine, specifically the development of SNOMED and CTV3. DATAMS creates a highly visual and intuitive interface for extracting anesthesia knowledge and representing it as a collection of terms and their relationships. It potentially solves a number of problems associated with the development of a data dictionary, including management and comparison of multiple disparate lists, creating a structure that allows the mapping of one term to another, and elimination of the need for everyone in anesthesia to use exactly the same terms when referring to a particular data element. This is an absolute necessity in the global practice of anesthesia where terms vary from country to country, among institutions within the same country, and among systems from different vendors."
The DDTF was first considered during a workshop on anesthesia information management systems held at the 2000 American Society of Anesthesiologists meeting. Subsequent discussions led to the formation of the DDTF and the development of a plan to review and compare the disparate data element lists that are part of every anesthesia information management system's (AIMS) computer-based patient record. An integral part of the DDTF's launch was the APSF's statement supporting the use of AIMS peri-operatively as a valuable resource to improve patient care and safety in anesthesia. The ability to capture and conduct research using standardized data from across AIMS as well as institutions is the ultimate outgrowth of the DDTF's activities. The anesthesia community has long been frustrated by the inability to collect and analyze uniform data elements across departments and institutions; a barrier to improving anesthesia care, identifying opportunities to reduce health care cost, and modifying practice appropriately.
The Task Force, a collaborative effort among manufacturers of anesthesia information management systems and the APSF, was charged with providing a series of "deliverables" (see box DDTF Deliverables). In April 2002, the group met to "spec out" and begin a development plan for the dataset. At that meeting, Dr. Sanderson presented a specification for a robust set of software tools that would allow the contribution and collection of multiple data lists, the mapping or comparison of those terms and the publication of a "key" that would allow toolset users to continue to name data elements flexibly, but allow the comparison of the underlying common meanings. Dr. Sanderson explained the concept of ontologies, or linguistic structures, on which other well-established systemized lexicons are built (e.g., SNOMED and National Health Service Clinical Terms Version 3 [NHS CTV3]), and their value in the DDTF situation where many existing data lists are being compared and the relationships between data elements established. At this point he proposed DATAMS to create, maintain, and disseminate a reference set of anesthesia terms and phrases. More technical information on DATAMS can be found here.
The group agreed that the DATAMS model was impressive, but asked Drs. Monk and Sanderson to attempt a more traditional approach utilizing lists submitted from vendor systems and spreadsheets to manage the data comparison. Over the Summer of 2002, the Clinical Working Group, the physician component of the DDTF, met frequently to compare data element lists from several institutions as well as those submitted from Picis, Draeger, eko, Philips, and Deio. Despite breaking the anesthesia record into more manageable sections such as patient information, business and regulatory terms, and equipment and the assistance of other clinicians such and Dr. David Reich and Dr. Homer Yang on the Working Group, Drs. Monk and Sanderson became convinced that the more structured DATAMS approach was the best path to follow. Further prompting their decision was the desire to provide meaningful results at the October 2002 ASA meeting so that the concepts and tools could be tested during the technical and scientific exhibitions.
From its inception, the DDTF sought to build on the experience of other medical specialties as well as similar initiatives within anesthesia. In May 2002, the group held its first serious discussions with SNOMED International, the developers of Systemized Nomenclature of Medicine (www.snomed.org). In addition, with the assistance of Dr. Martin Hurrell, the group also obtained a license for NHS CTV3. Dr. Yang joined the group in May as an official representative of the Canadian Anesthesia Society.
Discussions with SNOMED proceeded over several months and the Clinical Working Group evaluated its anesthesia content. The group agreed that while the structured approach taken by SNOMED and NHS CTV3 was sound, the anesthesia content relevant to North American practice was lacking. Sanderson and colleagues also evaluated the Apelon mapping tool that is used in the creation of SNOMED. Because the DDTF is committed to offering its reference data set freely to the anesthesia community, development using the SNOMED process seemed unlikely. The group decided to seek the services of experienced informatics linguists Mona and Murindar Singh to assist in the rapid development of a portion of the tools specified at the April meeting.
The results were presented at the ASA meeting and subsequently to STA and AACD. STA and AACD attendees were enthusiastic about the demonstrations provided by the DDTF. As a result, STA has offered to provide their list of clinical terms that formed the basis of a legacy data dictionary project and the AACD offered to contribute their Clinical Times Glossary to the database with appropriate attribution. Additional good news came from Raj Dash, MD, the Chair of the SNOMED Extensions Board, the body that approves additions to the SNOMED project, when he endorsed the DDTF's work.
The DDTF is on target with its plan to provide a mapping system to enable database conversions and bench mark studies across the databases of its supporting AIMS manufacturers. In addition, the flexible, open approach proposed by the group allows individual institutions and vendors to maintain distinct data terms or apply the most common terms uncovered through the DATAMS process. Most importantly the DDTF has secured the involvement and active participation of national and international professional societies and organizations. This data-oriented collaboration is a new support in the foundation on which the expansion of AIMS implementations can be based. The APSF is committed to encouraging the adoption of AIMS as a key to providing better anesthesia care and improved patient safety. The DDTF and DATAMS are helping to unlock that door.
For more information on the Data Dictionary Task Force, please see the DDTF section of the APSF's web site at www.apsf.org or contact the Project Director, Becky Haines, at ddtf@erols.com
The DDTF has been funded to develop specific items and deliver them to the anesthesia community in a form that can be used by vendor developers and anesthesia practitioners.
- A "Reference Set" of anesthesia terms
- Browse and download from an accessible website
- Provide vendor-specific mappings of their lists to the Reference Set
- Maintain a dynamic Reference Set
- Create a forum for the evolution of the lists
- Pursue the lists as a Standard under the aegis of APSF
|
