Rodney H Strahan1, Michal E Schneider-Kolsky. 1. Department of Diagnostic Imaging, Monash Medical Centre, Southern Health, Clayton, Victoria, Australia. rodney.strahan@southernhealth.org.au
Abstract
PURPOSE: Despite the frequent introduction of voice recognition (VR) into radiology departments, little evidence still exists about its impact on workflow, error rates and costs. We designed a study to compare typographical errors, turnaround times (TAT) from reported to verified and productivity for VR-generated reports versus transcriptionist-generated reports in MRI. METHODS: Fifty MRI reports generated by VR and 50 finalized MRI reports generated by the transcriptionist, of two radiologists, were sampled retrospectively. Two hundred reports were scrutinised for typographical errors and the average TAT from dictated to final approval. To assess productivity, the average MRI reports per hour for one of the radiologists was calculated using data from extra weekend reporting sessions. RESULTS: Forty-two % and 30% of the finalized VR reports for each of the radiologists investigated contained errors. Only 6% and 8% of the transcriptionist-generated reports contained errors. The average TAT for VR was 0 h, and for the transcriptionist reports TAT was 89 and 38.9 h. Productivity was calculated at 8.6 MRI reports per hour using VR and 13.3 MRI reports using the transcriptionist, representing a 55% increase in productivity. CONCLUSION: Our results demonstrate that VR is not an effective method of generating reports for MRI. Ideally, we would have the report error rate and productivity of a transcriptionist and the TAT of VR.
PURPOSE: Despite the frequent introduction of voice recognition (VR) into radiology departments, little evidence still exists about its impact on workflow, error rates and costs. We designed a study to compare typographical errors, turnaround times (TAT) from reported to verified and productivity for VR-generated reports versus transcriptionist-generated reports in MRI. METHODS: Fifty MRI reports generated by VR and 50 finalized MRI reports generated by the transcriptionist, of two radiologists, were sampled retrospectively. Two hundred reports were scrutinised for typographical errors and the average TAT from dictated to final approval. To assess productivity, the average MRI reports per hour for one of the radiologists was calculated using data from extra weekend reporting sessions. RESULTS: Forty-two % and 30% of the finalized VR reports for each of the radiologists investigated contained errors. Only 6% and 8% of the transcriptionist-generated reports contained errors. The average TAT for VR was 0 h, and for the transcriptionist reports TAT was 89 and 38.9 h. Productivity was calculated at 8.6 MRI reports per hour using VR and 13.3 MRI reports using the transcriptionist, representing a 55% increase in productivity. CONCLUSION: Our results demonstrate that VR is not an effective method of generating reports for MRI. Ideally, we would have the report error rate and productivity of a transcriptionist and the TAT of VR.