Table 2:
Author | Date | Title | SR or DT | Class 0 | Errors / Report | Class 1 | Errors / Report | Class 2 | Errors / Report | Class 3 | Errors / Report |
---|---|---|---|---|---|---|---|---|---|---|---|
Formatting Errors | Grammatically Incorrect | Meaning Different (error was obvious) | Meaning Different (error not obvious) | ||||||||
Basma et al. | 2011 | Error rates in breast imaging reports: comparison of automatic speech recognition and dictation transcription | SR | Punctuation error | 0.16 | Verb tense | 0.04 | Word omission | 0.14 | Wrong word | 0.01 |
Spelling mistakes | 0.03 | Plural | 0.06 | Word substitution | 0.13 | Incorrect measurement | 0.01 | ||||
Incomplete phrase | 0.01 | Nonsense phrase | 0.02 | Missing or added “no” | 0.01 | ||||||
Added word | 0.15 | ||||||||||
DT | Punctuation error | 0.02 | Verb tense | 0.02 | Word omission | 0.04 | Wrong word | 0.00 | |||
Spelling mistakes | 0.03 | Plural | 0.02 | Word substitution | 0.05 | Incorrect measurement | 0.00 | ||||
Incomplete phrase | 0.00 | Nonsense phrase | 0.00 | Missing or added “no” | 0.00 | ||||||
Added word | 0.04 | ||||||||||
Chang et al. | 2011 | Nonclinical errors using voice recognition dictation software for radiology reports: a retrospective audit | Combined | Type E – punctuation | 0.03 | Type F – other including spelling | 0.03 | Type A – wrong word substitution | 0.18 | Type C – deletion | 0.07 |
Type D – insertion | 0.13 | ||||||||||
SR | Type B – Nonsense phrase | 0.02 | |||||||||
DT | Type B – Nonsense phrase | 0.05 | |||||||||
David et al. | 2014 | Error rates in physician dictation: quality assurance and medical record production | SR | Made up words | 0.14 | Age mismatch | 0.03 | Wrong patient | 0.07 | ||
Gender mismatch | 0.05 | Wrong drug name/dosage | 0.05 | ||||||||
Wrong name | 0.01 | Name/dosage | 0.04 | ||||||||
Wrong doctor | 0.02 | Wrong lab values | 0.01 | ||||||||
Wrong date | 0.01 | Left/right discrepancy | 0.03 | ||||||||
Other | 0.20 | Medical discrepancy other | 0.00 | ||||||||
DT | Made up words | 0.11 | Age mismatch | 0.04 | Wrong patient | 0.05 | |||||
Gender mismatch | 0.02 | Wrong drug | 0.05 | ||||||||
Wrong name | 0.02 | Name/dosage | 0.01 | ||||||||
Wrong doctor | 0.02 | Wrong lab values | 0.03 | ||||||||
Wrong date | 0.03 | Left/right discrepancy | 0.02 | ||||||||
Other | 0.38 | Medical discrepancy other | 0.04 | ||||||||
McGurk et al. | 2008 | The effect of voice recognition software on comparative error rates in radiology reports | SR | Trivial (no changes to understanding) | 0.02 | Effect understanding | 0.03 | ||||
DT | Trivial (no changes to understanding) | 0.01 | Effect understanding | 0.01 | |||||||
Ramaswamy et al. | 2000 | Continuous speech recognition in MR imaging reporting: Advantages, disadvantages, and impact | SR | Irregular spacing | 2.98 | Spelling errors | 0.30 | Omissions and duplications | 0.372 | ||
DT | Irregular spacing | 0.11 | Spelling errors | 1.12 | Omissions and duplications | 0.1116 |
Class 0 errors: no change in meaning and grammatically correct; class 1 errors: no change in meaning but grammatically incorrect; class 2 errors: meaning was different but error was obvious; class 3 errors: meaning was different but not obvious.28