Method for inductive case-based decision support system over the Internet: exemplified by ear, nose and throat diagnostics in primary care
1997 (English)Doctoral thesis, comprehensive summary (Other academic)
This thesis is an example of an ear to loaf process:
A Preliminary Essential Data Set (PEDS) for Ear, Nose and Throat (ENT) diagnostics in Primary Health Care (PHC) was established by infologic modelling of existing tractates, between the specialist organisations of ENT and PHC, and four contemporary text-books on the subject (I). The PEDS was then compared to the aggregated data from a sample of case-files from a Swedish PHC centre. The quality of the individual medical record was found to be poor and not sufficient for follow-up and Quality Management (QM) purposes (II). The aggregated data from the case-file sample was processed by an inductive "expert" computer program that generated a decision-tree, based on the total number of decisions (diagnoses) in the data-base. This was nearly identical to the PEDS and somewhat less complicated. Infological modelling of the PEDS and the decision-tree resulted in the proposed Essential Data Set (EDS) for ENT diagnostics in PHC. The decision tree was "pruned" to achieve simplicity and ease of use (III). It was converted into a Computer-aided Decision Support System (CDSS) and put on the Internet using World Wide Web (WWW) tools (IV). This system was then implemented in the Skänninge PHC centre where an Electronic Medical Record System (EMRS) was in everyday use. Due to technical difficulties no integration between the two systems was possible. The CDSS was run on a computer outside the centre and this raised the question of secrecy and security (V). Prior to the implementation, and after a study period of four months attitudes towards, and the acceptance of, new Information Technology (IT) was explored by using focus group methodology (VI). It is important that those who are to use the system are intimately involved in its development. The results showed, that participation of the end-users during implementation is crucial for its acceptance. The need for a uniform and agreed terminology was obvious to facilitate co-operation and QM. The importance of nurses and physicians using the same terminology for documentation, was paramount.
Place, publisher, year, edition, pages
Linköping: Linköpings universitet , 1997. , 40 p.
Linköping University Medical Dissertations, ISSN 0345-0082 ; 538
Medical and Health Sciences
IdentifiersURN: urn:nbn:se:liu:diva-27562Local ID: 12224ISBN: 91-7219-083-3OAI: oai:DiVA.org:liu-27562DiVA: diva2:248114
1997-12-12, Administrationsbyggnadens aula, Universitetssjukhuset, Linköping, 09:00 (Swedish)
Papers, included in the Ph.D. thesis, are not registered and included in the posts from 1999 and backwards.2009-10-082009-10-082012-10-22Bibliographically approved