The purpose of this study was to examine structured template use among emergency radiologists, and if this influences audio dictation time, radiology report length, or total radiologist study time. Retrospective data collection of consecutive occurrences of seven common imaging examinations interpreted by a dedicated emergency radiology division over a 2-month period yielded 3449 reports. Templates had been in place for >3 years. For each examination, we documented the individual audio dictation time (ADT), total words, and total time the radiologist spent on a study from report creation until final signing. In 81.2 % (n = 2772) of all cases, a basic template was used. In 2.8 % (n = 78) of these template-use cases, the radiologist removed key elements from the structured template. Of the 3417 reports with complete data, mean ADT was 37.3 s, mean word length was 132.3 (of which, on average, 64 were dictated), and total radiologist time per study (TRT) was 349.7 s. Study type was significantly associated with ADT, total words, and TRT (p < 0.001). Template usage decreased ADT (p < 0.001) by 47 %, but did not affect total word length or TRT. Parameters varied by individual attending (p < 0.001): 20 % (2/10) of attendings had differences in report length when using versus not using templates (p < 0.001). With long-term template usage, compliance with structured templates is high, and few radiologists significantly alter the templates. Template use decreases ADT and for a small fraction of radiologists impacts total word length and has a mixed impact on TRT.
Keywords: Informatics; Radiology; Speech recognition; Structured report; Template; Turnaround time.