Speech recognition technology and transcription: Build quality assurance into your program
Speech recognition technology coupled with transcription services has many benefits, including saving providers time and money. But not all companies assure the quality of their products and services, which can create serious problems for organizations.
A recent article published on The Joint Commission’s website highlights some of the vulnerabilities of speech recognition technology (SRT), which include improper use and expectations of SRT capabilities, mismanagement of SRT with degradation of translation over time, no standards for style, grammar and readability, and unclear roles and standards for editors of transcribed medical notes.
Additionally, the article points out, there are insufficient comprehensive quality assurance and process improvement programs for health care documentation. The result is lack of accountability, minimal regulatory oversight, and almost no monitoring of documentation processes.
Surveys show that critical error rates for speech recognition proofreading and editing are as high as 22 percent, according to an article published in the Journal of the American College of Radiology. It has been found that voice recognition takes 50 percent longer to dictate and there are 5.1 errors per case. Additionally, 90 percent of all voice recognition dictations contained errors prior to physician sign off. After sign off, 35 percent of voice recognition still had errors, reports the Journal of Digital Imaging.
What’s more, a recent study concluded that the physician-as-editor model does not guarantee that error will be found, and that physicians do not always take the time to edit and proofread the transcribed medical record, according to an article published in the International Journal of Health Care Quality Assurance.
So what can your organization do to minimize its risk of serious documentation errors?
To get the most benefit from SRT, AHIMA advises implementing consistent policies and procedures to address dictation best practices. This includes a style guide and consistency when applying edits.
AHIMA also recommends third-party editing. In current transcription practices, many reports are not reviewed closely before physicians sign them. Organizations may want to consider implementing a quality assurance program for physician front-end speech created documentation in addition to their current transcription quality assurance program.
Finally, AHIMA suggests that each organization define its acceptable standards of accuracy for all documentation, whether it is handwritten, checked off a form, dictated as free text, dictated for processing by speech recognition (front-end or server), or entered into an EHR by keyboard or speech commands.
Photo credit: Tnarik Innael (cc)