Achim Escher, Presenter: Nothing to Disclose

Tobias Heye MD, Abstract Co-Author: Nothing to Disclose

Elmar M. Merkle MD, Abstract Co-Author: Advisor, Siemens AG Advisor, Bayer AG Speakers Bureau, Bayer AG Reserach Support, Bayer AG Reserach Support, Bracco Group Reserach Support, Guerbet SA

Despite a substantial reduction in report TAT in 2013 there was no evidence of a negative impact on the structural report quality. This may be explained by the assumption that TAT improvement is a matter of optimizing time efficiency and TAT awareness rather than shortening the time available for reporting.

Radiology reports from a time period before reduction in TAT (06-12/2012) were compared to reports from a time period with significant reduction in TAT (06-12/2013). Plain text data from 1’812 radiology reports containing the keywords “pulmonary embolism” (PE) and “cerebral ischemia” (CI) were retrieved from the radiology information system. An advanced language processing approach was used to analyze the following objective criteria for report quality: word count and its variability, the number of addendums to reports and the adherence to common reporting structures (history, procedure, comparison, findings, impression). It was also assessed if the impression addressed the clinical question which initiated the procedure.  

No significant difference in mean word count (CI: 108.3 to 127.7 words; PE: 125.4 to 125.4 words) and variability (standard deviation; CI: 48.0 to 58.7; PE: 65.4 to 64.2) was found. The number of addendums slightly increased but remained on a very low level in absolute numbers (CI: 0.8% to 1.5%; PE: 0.6% to 1.6%). The adherence to common reporting structures showed a slight improvement in some categories (e.g. reference to comparison studies, case history). 98.0% of reports addressed PE as clinical question in 2012, which changed to 97.8% in 2013 (CI: 73.0% to 66.1% respectively). The percentage of reports addressing PE in the line in the impression decreased from 69.3% to 65.0% (CI: 28.6% to 27.6% respectively).

The approach of analyzing structural report quality can be used to detect changes in the report quality. In this work we could show that improving report TAT does not negatively impact report quality. Although this approach does not measure the clinical accuracy of reports, it offers objective measures of the care that goes into reporting.

Escher, A, Heye, T, Merkle, E, Faster Report Turnaround-times versus Report Quality! Two Opposing Objectives?.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14045657.html