JOURNAL CLUB: Voice Recognition Dictation: Analysis of Report Volume and Use of the Send-to-Editor Function

Abstract

The purpose of this study was to evaluate use of the send-to-editor function of a radiology voice recognition dictation system and compare study volumes of radiologists who self-edit with those of radiologists who send reports to the editor. Use of voice recognition shortcuts was also evaluated. Voice recognition dictation systems were installed in a six-hospital system, including an 800-bed tertiary care center and five community hospitals, in 2002. This became the only means of radiologist dictation in July 2005. Report volumes, use of the send-to-editor function, and use of shortcuts were tracked from October 2005 through October 2008. A subspecialty private radiology group, ranging from 37 radiologists in July 2005 to 50 radiologists in October 2008, interpreted the imaging studies. Radiologists had no financial incentives to self-edit. The percentage of radiologists using the send-to-editor function remained relatively constant at 46%, resulting in 21% of total reports sent to the editor. Radiologists who used the send-to-editor function dictated approximately 41% more reports than those who self-edited. The volume of reports generated by general radiologists reading large volumes of computed radiography cases and sending to the editor was greater than that of radiologists who self-edited (p < 0.05). There was no significant difference between radiologists who self-edited and those who sent to the editor with respect to number of shortcuts used. Radiologists reading large volumes of computed radiography cases and using the send-to-editor function generated significantly more reports than radiologists who did not, suggesting that the send-to-editor function may be useful for improving productivity among radiologists reading large volumes of computed radiography cases.