Affiliation: | a Department of Population Medicine, Ontario Veterinary College, University of Guelph, Guelph, ON N1G 2WJ, Canada b Department of Clinical Studies, Ontario Veterinary College, University of Guelph, Guelph, ON N1G 2W1, Canada c Information Technology Services, Ontario Veterinary College, University of Guelph, Guelph, ON N1G 2W1, Canada |
Abstract: | This study was designed to evaluate the quality of data from computerized medical record abstracts at the Veterinary Teaching Hospital, Ontario Veterinary College. Information in the paper medical record (registration, physical exam, daily progress, laboratory, radiology, anaesthesia and surgery forms), the summary sheet, and the computerized record were compared. A random sample of 100 patient visits from a subset of visits that were identified as all possible dog or cat elective-surgery visits in the computerized records from 1983 through 1991 were used. Clinicians were responsible for summarizing the medical record on the summary sheet and health record technicians entered the information from the summary sheet into the computer record. Most of the discrepancies (n = 33) noted for diagnoses, procedures or complications were due to the lack of transfer of information from the paper forms to the summary sheet. Although the medical record technicians detected and entered some of these (n = 10), most were not entered into the computerized records. The next most-common discrepancies (n = 5) were due to the clinician writing a different entry on the summary sheet than was present on the other paper record forms. Only one record had a miscoded diagnosis and three had miscoded procedures. Misfiled forms in the medical record folders of two patients resulted in one incorrect diagnosis and one incorrect procedure recorded in the computer record. When any discrepancy between the paper and computer record for diagnoses, procedures or complications was considered, 41% of the visits had discrepancies (46% dogs; 36% cats). The percentage varied by surgery status: non-elective surgeries 82%, elective surgeries 37% and non-surgical visits 29%. There was no obvious time trend. The completeness and accuracy of information in the database was inadequate for the intended research on post-operative complications following elective surgeries. |