Voice Recognition and the Electronic Health Record

Voice recognition technology is replacing conventional dictation in many healthcare information systems, Electronic Health Record (EHR) included. Voice recognition is certainly able to eliminate transcription costs, but how about transcription errors? Is it able to listen and interpret better than a human?

The answer to both questions is, yes, especially if it is “trained.” A physician can generally start using voice recognition and expect 95 percent accuracy.

The natural language processing characteristics of voice recognition technology allow spoken words to be broken down into specific data fields, not just blocks of free text. Voice recognition can be highly intuitive, if an EHR system is programmed to incorporate dynamic, command-based responses.

If an EHR system is meant to function jointly with voice recognition technology, physicians should not have to speak in complete sentences or provide comprehensive narratives. An EHR system can, and should, be programmed to exercise dynamic, command-based responses consistent with specific types of procedures, techniques, symptoms, care plans, etc.

Thousands of dynamic, command-based responses programmed within an EHR system can substantially reduce the time it would take to perform conventional dictation. Also, physicians’ voice files, which they can train using their voice and make corrections in real time to the text in the EHR, can now be stored in the cloud. This means physicians can access the same voice files whether they’re documenting in the EHR during a patient visit or on their mobile phone. Plus, the need for transcription is removed entirely, easily saving the average physician $30,000 to $50,000 a year.

Trained voice recognition also helps overcome many of the issues surrounding general dissatisfaction with EHR systems. In the absence of voice recognition, physicians usually encounter a lengthy series of screens, tabs, check boxes, etc. This causes them to exhaust 5 to 12 minutes, more than 100 mouse clicks, and an abundance of manual data entry to produce a single exam note. Physicians are frustrated because they are not able to have enough face time with patients, instead, they’re spending too much time doing data entry and not seeing as many patients. Or they’re spending time doing data entry after the work day. With trained voice recognition and dynamic, command-based responses, a single exam note should take less than 90 seconds, and can also help with preventing physician burnout.

By adopting an EHR with trained voice recognition, a physician practice can save a substantial amount of time and money, and increase the number of patients they attend to.

Money talks and voice recognition listens.


Have questions? Leave us a comment and let us know!


Leave a Reply