Speech recognition for clinical documentation from 1990 to 2018: a systematic review

Abstract

Objective The study sought to review recent literature regarding use of speech recognition (SR) technology for clinical documentation and to understand the impact of SR on document accuracy, provider efficiency, institutional cost, and more. Materials and Methods We searched 10 scientific and medical literature databases to find articles about clinician use of SR for documentation published between January 1, 1990, and October 15, 2018. We annotated included articles with their research topic(s), medical domain(s), and SR system(s) evaluated and analyzed the results. Results One hundred twenty-two articles were included. Forty-eight (39.3%) involved the radiology department exclusively and 10 (8.2%) involved emergency medicine; 10 (8.2%) mentioned multiple departments. Forty-eight (39.3%) articles studied productivity; 20 (16.4%) studied the effect of SR on documentation time, with mixed findings. Decreased turnaround time was reported in all 19 (15.6%) studies in which it was evaluated. Twenty-nine (23.8%) studies conducted error analyses, though various evaluation metrics were used. Reported percentage of documents with errors ranged from 4.8% to 71%; reported word error rates ranged from 7.4% to 38.7%. Seven (5.7%) studies assessed documentation-associated costs; 5 reported decreases and 2 reported increases. Many studies (44.3%) used products by Nuance Communications. Other vendors included IBM (9.0%) and Philips (6.6%); 7 (5.7%) used self-developed systems. Conclusion Despite widespread use of SR for clinical documentation, research on this topic remains largely heterogeneous, often using different evaluation metrics with mixed findings. Further, that SR-assisted documentation has become increasingly common in clinical settings beyond radiology warrants further investigation of its use and effectiveness in these settings.

Essential Elements of Natural Language Processing: What the Radiologist Should Know

“…The implementation of ASR in clinical workflows are typically aimed at productivity improvements and cost reduction, more widely implemented in radiology than in other medical fields. A recent systematic review found that approximately 40% of the academic publications in ASR for clinical documentation have been published in radiology (47). Among these, the most studied measure of ASR is in the time for documentation, where the results have been mixed À some studies reveal a decrease in overall documentation time, while others conclude an increase (48,49).…”

Advancing biomedical and health informatics knowledge through reviews of existing research

“…The purpose may be relatively narrow or broad in scope. For example, Koleck et al 8 focused on use of natural language processing to process or analyze information related to a selected set of symptoms in electronic health record free-text narratives while other systematic reviews examined broader topics such as clinical pathways, 9 speech recognition technology for clinician documentation, 10 and evaluation approaches for visual analytic technologies in health. 6…”

The impact of implementing speech recognition technology on the accuracy and efficiency (time to complete) clinical documentation by nurses: A systematic review

“…Furthermore, SRT developers claim that the software can recognise users and their accents, easily update and individualise patient records, and synchronise it with other medical records of the patient, capturing data even from noisy environments (Tpro.io, 2013). Newer systems also “Learn” individual user's natural speech patterns, thereby progressively reducing word‐error rates with subsequent usage (Blackley, Huynh, Wang, Korach, & Zhou, 2019). However, notable differences have been reported by clinicians between different SRT softwares and different clinical environments, which could potentially affect documentation quality by impeding or enhancing the end result (Behan, 2018).…”