Longitudinal Comparative Study on the Influence of Computers on Reporting of Clinical Data

Abstract

The impact of the clinical database system SISCOPE on medical services was evaluated and objective data compiled on the quality of information recording and reporting using a fully structured data entry system compared to traditional free text reporting. 1565 upper endoscopy reports produced with SISCOPE over a period of 12 months were assessed for completeness and compared to 152 and 208 free text reports done 4 months before and 1 month after the study period, respectively. Data on four common gastrointestinal findings (esophageal varices, ulcers, polyps and tumors) were evaluated. Physicians' compliance with the new system was good, as reflected by a constant level of quality of reporting over time, although a very slight decline in the ratio of computer generated reports to the total number of examinations was noted. Structured reports had an 18% missing data rate and contained 60% more relevant information than free text reports, which had a 48% missing data rate. No educational effect of the system was seen as missing data rates returned to pre-computerization levels just one month after the end of the study. It is concluded that menu-driven structured data entry systems result in production of far superior reports as compared to free text systems, probably due to their reminder effect.

Compliance with terminology standards in reflux, ulcers, and gastritis: A study of 881 consecutive upper gastrointestinal endoscopy reports

“…For the diagnosis and morphological description of ªgastritisº or a normal endoscopy, the results were even more inconsistent: According to the consensus terminology, the term ªgastritisº should not be used, and descriptive features such as erythema, erosions, etc., are recommended instead. This is based on the fact that previous studies showed a weak correlation between endoscopic signs of ªgastritisº and the 1 Circular, tongues, combination of the two 2 < 3 cm, > 3 cm 3 Location: a) sections of the stomach (cardia, fundus, body, antrum, prepyloric; proximal and distal bulb; b) parts of the circumference (anterior and posterior wall, greater and lesser curvature) 4 Circumscript, patchy, linear, diffuse 5 Flat or elevated (papule) erosions 6 Fibrinous, hematogenous, combination of the two, newly epithelialized Table 2 Completeness of the descriptions of the various parameters of the esophageal diagnoses listed below when compared to the German standard terminology (n = 516 endoscopy reports analyzed); 304 reports described normal findings. .…”

“…However, there has been continuing debate on whether adherence to this type of standard terminology is likely to become everyday routine in clinical practice. Previous reports have shown that up to 50 % of the relevant parameters (e. g., the size of a lesion) are missing in routine endoscopy reports produced using free dictation [1,4,5]. The aim of the present study was to analyze, in a retrospectively selected sample of upper gastrointestinal endoscopy reports collected at various centers over a one-month period, the extent to which the diagnoses and descriptive terms used were in concordance with a standard terminology, which was developed at the same time and approved thereafter by the Germans Society (DGVS) [3].…”

Development, Implementation, and Evaluation of a Structured Reporting Web Tool for Abdominal Aortic Aneurysms

“…Such applications have been developed, used, and also associated with accurate and complete data entry. The improvement and completeness of structured reports compared with free-text reporting are proved [23,24]. For pelvic ultrasound, a structured data entry system named UltraSTAR has been implemented to reach a high satisfaction among radiologists and gain slightly more complete but certainly more structured data than free-text reports [25].…”

Endoscopic electronic medical record systems

“…One study indicated that menu-driven, structured, data-entry systems resulted in fewer reports with missing data compared with free text reporting (18% of reports vs 48%). 12 Another study demonstrated that the time needed to generate an endoscopy report was similar whether generated by handwritten (113 seconds; range, 108-117), dictated (including transcriptionist's time) (237 seconds; range, 225-250), or computer-generated means (102 seconds; range, 95-110). 13 A Dutch group showed that by using text blocks based on anatomic landmarks and diagnoses, an endoscopist could generate 90% of reports within 2 minutes.…”