Doctors' poor handwriting is a well-recognized source of medical errors, but their sometimes sloppy speech habits are a less well-known contributor.
A study based on a sample of 220 dictated medical records totaling 9,726 lines of transcription found 27% of the 96 more-serious flaws were attributed to the speaker, typically a physician, and not the transcriptionist. Twenty of 38 critical flaws (53%) and six of 58 major flaws (10%) were traced to the speaker.
Critical flaws include patient misidentification, medical word misuse and omitted dictation. Major flaws include misspellings and inappropriate blanks. Minor flaws include punctuation, grammar and formatting errors.
Judy Hinickle, owner of TransCom Corp., a Menomonee Falls, Wis.-based medical transcription and consulting company, developed the software program, QA Navigator, which scored the sample against the Metrics for Measuring Quality in Medical Transcription, a set of quality standards being developed by the American Association for Medical Transcription, an industry trade group.
Physician-contributed dictation errors are an acknowledged problem in the transcription industry. Part of the trouble arises when busy doctors engage in multitasking, attempting dictation while doing other things, such as eating.
"Transcriptionists call it chew and slurp," says Frank Lavelle, president of medical transcription service organization MedQuist. But it sometimes goes beyond a simple breach of etiquette.
"If you've ever received dictation coming from a surgeon driving down the road in an open convertible eating an apple and dictating a report for a procedure two weeks ago that he's just now getting around to, that's hard," says David Woodrow, a division head of SPI Technologies, also a medical transcription service organization.
AAMT Executive Director Peter Preziosi sent Hinickle's findings to the Joint Commission on Accreditation of Healthcare Organizations with an eye to including transcription errors in its sentinel-events reporting program.
Richard Croteau, executive director for patient-safety initiatives for the JCAHO, says he was "somewhat surprised" the results "did not show even more of the problems originating with the" speaker.
Croteau says he supports standard-setting for transcription and "continued monitoring of the accuracy of the voice-to-text conversion process" to improve accuracy and reliability. While there is no data on how often dictation or transcription problems contribute to adverse patient outcomes, "I am persuaded they are a significant factor," he says.