Establishment of methods for extracting and analysing patient data from electronic practice management software systems used in first opinion veterinary practice in the UK

Jones-Diette, Julie Susan (2014) Establishment of methods for extracting and analysing patient data from electronic practice management software systems used in first opinion veterinary practice in the UK. PhD thesis, University of Nottingham.

[img]
Preview
PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (4MB) | Preview

Abstract

Examining patient records is a useful way to identify common conditions and treatment outcomes in veterinary practice and data gathered can be fed back to the profession to assist with clinical decision making. This research aimed to develop a method to extract clinical data from veterinary electronic patient records (EPRs) and to assess the value of the data extracted for use in practice-based research. The transfer of new research from continuing professional development (CPD) into practice was also considered.

An extensible mark-up language (XML) schema was designed to extract information from a veterinary EPR. The analysis of free text was performed using a content analysis program and a clinical terms dictionary was created to mine the extracted data. Data collected by direct observation was compared to the extracted data. A review of research published in the proceedings of a popular veterinary CPD event, British Small Animal Veterinary Association (BSAVA) Congress, was appraised for evidence quality.

All animal records were extracted and validation confirmed 100% accuracy. The content analysis produced results with a high specificity (100%) and the mined data analysis was successful in assessing the prevalence of a specific disease. On comparison, the data extracted from the EPR contained only 65% of all data recorded by direct observation. The review of BSAVA Congress abstracts found the majority of the clinical research abstracts (CRAs) presented to be case reports and case series, with differences in focus between CRAs and veterinary lecture stream abstracts.

This study has demonstrated that data extraction using an XML schema is a viable method for the capture of patient data from veterinary EPRs. The next step will be to understand the differences found between data collected by observation and extraction, and to investigate how research presented as CPD is received, appraised and applied by the veterinary profession.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Dean, R.
Brennan, M.
Cobb, M.A.
Subjects: R Medicine > R Medicine (General)
S Agriculture > SF Animal culture
Q Science > QA Mathematics > QA 75 Electronic computers. Computer science
Faculties/Schools: UK Campuses > Faculty of Medicine and Health Sciences > School of Veterinary Medicine and Science
Item ID: 14345
Depositing User: EP, Services
Date Deposited: 03 Oct 2014 09:29
Last Modified: 13 Sep 2016 21:39
URI: http://eprints.nottingham.ac.uk/id/eprint/14345

Actions (Archive Staff Only)

Edit View Edit View