Challenges in the development and testing of a reference terminology model for nursing interventions.

The purpose of this paper is to report the work of the Interventions Group of the first Nursing Terminology Summit (1999) and to describe the challenges and insights of this group as they have evolved a reference terminology model for nursing interventions. When the group began its work as part of the first meeting of the Nursing Terminology Summit, it had the overall objective of proposing the intervention component of a reference terminology model for nursing. Although there is not a definitive proposal for this to date, the group's exploration and analysis has clarified and explicated both the types of models of clinical information and the current " state of the art" of formal representations of nursing interventions as well as the relationship of nursing languages to these formal representations. In addition, it is our perspective that the work of this group is representative of the process and challenges facing many similar groups currently engaged in modeling efforts. Consequently, critical success factors of such efforts are identified and discussed. This paper reports both the specific outputs of the group related to progress in defining a terminology model of nursing interventions and observations and lessons learned regarding consensus modeling work.

Integrating medical information and knowledge in the HL7 RIM.

Guidelines have a proven ability to improve quality of health care and reduce cost, yet, guidelines are not very well deployed at the point of care. This is largely due to the impedance mismatch between decision support modules and the Electronic Medical Record (EMR.) The Unified Service Action Model (USAM) as part of the HL7's Reference Information Model provides a conceptual integration between patient data and medical knowledge. The USAM defines one action-oriented information structure for patient data, concept definitions, action plans, conditionals, and goals. This suggests a new approach to the problem of sharing data and knowledge, effectively working around the problem of missing domain terminology.

Challenges and issues related to implementation of nursing vocabularies in computer-based systems.

As key stakeholders from the clinical setting and vendor communities, the authors share a summary of their collective experience related to the challenges and issues associated with implementing the vocabularies recognized by the American Nurses Association in several installations of commercially available clinical information systems. Although the focus of the article is on summarizing the challenges and issues, it is of note that the authors' experiences across care settings suggest that the experience and effort of using one of the ANA-recognized vocabularies in a computer-based system are essentially worthwhile and positive. The issues and challenges fall into two categories: 1) those related to the developmental status of nursing vocabularies, and 2) those related to the adoption or implementation of new technology.

Nursing classification systems: necessary but not sufficient for representing "what nurses do" for inclusion in computer-based patient record systems.

Our premise is that from the perspective of maximum flexibility of data usage by computer-based record (CPR) systems, existing nursing classification systems are necessary, but not sufficient, for representing important aspects of "what nurses do." In particular, we have focused our attention on those classification systems that represent nurses' clinical activities through the abstraction of activities into categories of nursing interventions. In this theoretical paper, we argue that taxonomic, combinatorial vocabularies capable of coding atomic-level nursing activities are required to effectively capture in a reproducible and reversible manner the clinical decisions and actions of nurses, and that, without such vocabularies and associated grammars, potentially important clinical process data is lost during the encoding process. Existing nursing intervention classification systems do not fulfill these criteria. As background to our argument, we first present an overview of the content, methods, and evaluation criteria used in previous studies whose focus has been to evaluate the effectiveness of existing coding and classification systems. Next, using the Ingenerf typology of taxonomic vocabularies, we categorize the formal type and structure of three existing nursing intervention classification system--Nursing Interventions Classification, Omaha System, and Home Health Care Classification. Third, we use records from home care patients to show examples of lossy data transformation, the loss of potentially significant atomic data, resulting from encoding using each of the three systems. Last, we provide an example of the application of a formal representation methodology (conceptual graphs) which we believe could be used as a model to build the required combinatorial, taxonomic vocabulary for representing nursing interventions.

Health status measurement in computer-based patient record systems.

As requests for information about quality of health care have increased, purchasers of computer-based patient record (CPR) systems are demanding information regarding the manner in which these systems can assist in measuring health care quality. The type of information requested by purchasers of health care, accrediting agencies, and consumer groups is shifting toward more complex measures of health status. Both provider and patient perceptions of health status are relevant in measuring the impact of health care interventions. Standardized coding and classification systems and standardized health status measurement instruments each offer utility in capturing and representing health status in CPR systems.

Documenting 'what nurses do'--moving beyond coding and classification.

A variety of strategies for knowledge representation have been applied to the texts from a number of medical domains. Many of the techniques rely on the well-defined ways in which medical terms are used within a given domain, a phenomenon referred to as 'sublanguage.' Because much of nursing documentation involves the use of 'everyday' language, the viable application of sublanguage-based approaches to knowledge representation of nursing documentation is not a forgone conclusion. We propose an approach utilizing semantic markup of nursing notes as a strategy for determining whether the documentation of 'what nurses do' is a sublanguage Results of an initial feasibility study utilizing the approach are presented.

Evaluating standardized coding and classification systems for clinical practice: a critical review of the nursing literature in the United States.

Developers of healthcare information systems are challenged by the difficulty of meeting the simultaneous goals of 1) capturing and electronically representing the broad array of data related to healthcare with sufficient expressibility to provide adequate documentation of the patient encounter, and 2) utilizing standardized coding and classification systems to facilitate linkages among computer-based systems. The evaluation studies to date have primarily focused on matching actual clinical data with terms in the recognized classification systems. These studies have provided evidence that the classification systems are relatively domain complete for categorizing patient problems and nursing interventions. Although several of the published criteria for evaluation relate to structure, it is noteworthy that this has not yet been a major focus of study in nursing. There are several areas of critical need that must be addressed. First, additional work is needed to develop and refine a standardized set of atomic-level terms relevant to nursing, including those for assessments, problems, and activities. Second, knowledge representations must be developed to support the building of complex concepts from atomic-level data.

Standardized nursing classification systems: necessary, but not sufficient, for representing what nurses do.

The American Nurses Association Steering Committee on Databases to Support Nursing Practice has recognized three standardized nursing intervention classification systems. Because each of these classifications systems focuses on encoding informational abstractions of nursing actions, rather than providing a controlled vocabulary and compositional grammar from which informational concepts and abstractions can be constructed, the systems are necessary, but not sufficient, for representing what nurses do. In particular, computer based patient record systems focused on process understanding and process improvement will require atomic-level representations of nursing actions suitable for transformation into a variety of information abstractions, including, but not limited to, the abstractions contained in the three existing classification systems.

Stochastic simulation algorithms for query networks.

One of the barriers to using belief networks for medical information retrieval is the computational cost of reasoning as the networks become large. Stochastic simulation algorithms allow one to compute approximations of probability values in a reasonable amount of time. We previously examined the performance of five stochastic simulation algorithms applied to four simple belief networks networks and found that the Self-Importance algorithm performed well. In this paper, we examine how the same five algorithms perform when applied to a belief network derived from the cardiovascular subtree of the Medical Subject Headings (MeSH). Both the Likelihood Weighting and Self-Importance algorithms perform well when applied to the MeSH-derived network, suggesting that stochastic simulation algorithms may provide reasonable performance in medical information retrieval settings.

A detection algorithm for multiform premature ventricular contractions.

This paper reports an algorithm developed to identify and quantify multiform PVCs. The algorithm clusters PVCs of similar morphology using a combination of time-domain and frequency-domain analysis. Initially, PVCs are grouped together on the basis of four time-domain-based morphological feature measurements. However, these time-domain-based clusters many times are nonunique because commonly encountered signal changes can cause substantial variations in the feature measurements of clinically similar beats. These redundant clusters are consolidated using two frequency-domain parameters: The First Spectral Moment (FSM) (center of gravity) of the amplitude spectrum, and the 5-Hz phase angle.