(Taken from application abstract): Nursing documentation consists of primarily narrative text in patient records. The clinical data that reside in this narrative are currently not easily retrievable for use in clinical studies, or for quality improvement. The immature scientific state of nursing's clinical vocabulary dictates that automated methods to make this data useful be explored, so that these rich clinical nursing data are available for use. The overall goal is to demonstrate that information about home care nursing relevant to statistical and database analysis of nursing practice can be extracted automatically from nursing narrative; that these data can be used for classification; and, that these automated methods are extensible to other home care nursing record domains and standard nursing terminologies. Specifically, we propose to: (1) Explore the ability of recently developed statistically-based linguistic methods to accurately assign part-of-speech tags to and parse (i.e., assign syntactic structure to) SOAIP note documentation of home care nurses. Part of speech tagging can reduce lexical ambiguity, while parsing can reveal the argument and complement structure of verbs, which can then be used to focus on evidence relevant to categorizing narrative descriptions and translating them into a standard terminology. (2) Explore the ability of widely available software tools and machine-readable linguistic resources to summarize, in terms of categories from Grobe's Nursing Intervention Lexicon and Taxonomy (NILT)(Grobe, 1995), the interventions in the """"""""I"""""""" (Intervention) section of the tagged and parsed SOAIP notes. (3) Explain the domain portability of these methods by applying them to another domain of home care nursing record narratives. (4) Explore the task extensibility of these methods by applying them to the task of translating tagged and parsed descriptions of nursing interventions into phrases from one or more of the UMLS Metathesaurus standard nursing terminologies, such as the Omaha interventions (Martin, 1995), Saba's Home Health Care Classification (HHCC)(Saba, 1995), or the Nursing Interventions Classification (NIC) (Bulechek, McCloskey & Donahue, 1995).
| Cawsey, A J; Webber, B L; Jones, R B (1997) Natural language generation in health care. J Am Med Inform Assoc 4:473-82 |
