Showing papers by "James J. Cimino published in 1997"

Usability testing in medical informatics: cognitive approaches to evaluation of information systems and user interfaces.

Abstract: This paper describes an approach to the evaluation of health care information technologies based on usability engineering and a methodological framework from the study of medical cognition. The approach involves collection of a rich set of data including video recording of health care workers as they interact with systems, such as computerized patient records and decision support tools. The methodology can be applied in the laboratory setting, typically involving subjects "thinking aloud" as they interact with a system. A similar approach to data collection and analysis can also be extended to study of computer systems in the "live" environment of hospital clinics. Our approach is also influenced from work in the area of cognitive task analysis, which aims to characterize the decision making and reasoning of subjects of varied levels of expertise as they interact with information technology in carrying out representative tasks. The stages involved in conducting cognitively-based usability analyses are detailed and the application of such analysis in the iterative process of system and interface development is discussed.

Supporting infobuttons with terminological knowledge.

Abstract: We have developed several prototype applications which integrate clinical systems with on-line information resources by using patient data to drive queries in response to user information needs. We refer to these collectively as infobuttons because they are evoked with a minimum of keyboard entry. We make use of knowledge in our terminology, the Medical Entities Dictionary (MED) to assist with the selection of appropriate queries and resources, as well as the translation of patient data to forms recognized by the resources. This paper describes the kinds of knowledge in the MED, including literal attributes, hierarchical links and other semantic links, and how this knowledge is used in system integration.

Using Logical Observation Identifier Names and Codes (LOINC) to exchange laboratory data among three academic hospitals.

Abstract: Using a standard set of names and codes to exchange electronic laboratory data would facilitate multiinstitutional research and data pooling. This need has led to the development of the Logical Observation Identifier Names and Codes (LOINC) database and its test naming convention. We conducted a study which required 3 academic hospitals (in 2 separate medical centers) to extract raw laboratory data from their local information system for a defined patient population, translate tests into LOINC, and provide aggregate data which could then be used to compare laboratory utilization. We found that the coding of local tests into LOINC can often be complex, especially the "Kind of Property" field, and apparently trivial differences in choices made by individual institutions can result in nonmatches in electronically pooled data. In our study, 72-86% of the failures of LOINC to match the same tests between different institutions were due to differences in local coding choices. LOINC has tremendous potential to eliminate the needing for detailed human inspection during the pooling of laboratory data from diverse sites, and perhaps even a built-in capability to adjust matching stringency by selecting subsets of LOINC fields required to match. However, a quality, standard coding procedure at all sites is critical.

Controlled vocabulary and design of laboratory results displays.

Abstract: Traditional data-review displays are driven by the ancillary systems that produced the data. A different paradigm is being used at Columbia-Presbyterian Medical Center (CPMC) where a controlled medical vocabulary-the Medical Entities Dictionary (MED) is the driving force behind laboratory data-review displays. Using hierarchical and semantic networks the authors have constructed a Web-based tool that considerably simplifies the MED-editing task required to create new displays. The tool uses knowledge in the MED to extract contextually relevant hierarchic and semantic sub-nets from the MED. The tool has a sensitivity of 92.2% and a relevance of 94.7% for retrieval of terms from the MED. Based on these results and given sufficient domains' structure within controlled vocabularies, we conclude that similar algorithms will enable applications to design and generate customized displays on-the-fly.

Linking a clinical system to heterogeneous information resources.

Abstract: We present a model for providing clinical information system (CIS) users with quick access to high quality information resources. We have developed a chest X-ray information button application which is attached to the chest X-ray reports in the CIS at the Columbia Presbyterian Medical Center. The application generates questions based on clinical information, user interest, generic question templates, and resource availability. It then provides answers to the questions through integrated access to heterogeneous information resources including the CIS itself and publicly accessible Web resources.

Partitioning a vocabulary's IS-A hierarchy into trees.

Abstract: Controlled medical vocabularies are useful in application areas such as medical information-systems and decision-support. However, such vocabularies are large and complex, and working with them can be daunting. It is important to provide a means for orienting users to the vocabulary's contents. This paper introduces a methodology for partitioning a vocabulary into small, meaningful pieces. The partitioning is done with respect to the vocabulary's IS-A hierarchy. The methodology, based on a set of rules for refining the IS-A hierarchy, is a process carried out by a user in conjunction with the computer. The methodology is demonstrated on a complex portion of a vocabulary.

Intranet technology in hospital information systems.

Abstract: The clinical information system architecture at the Columbia-Presbyterian Medical Center in New York is being incorporated into an intranet using Internet and World Wide Web protocols. The result is an Enterprise-Wide Web which provides more flexibility for access to specific patient information and general medical knowledge. Critical aspects of the architecture include a central data repository and a vocabulary server. The new architecture provides ways of displaying patient information in summary, graphical, and multimedia forms. Using customized links called Infobuttons, we provide access to on-line information resources available on the World Wide Web. Our experience to date has raised a number of interesting issues about the use of this technology for health care systems.