SpeechCheck is...

the next step in the speech recognition/self-editing process. Our goal is to take the training given by the speech recognition vendor one step further - to train your physicians on the best ways to dictate and edit, with a goal of an overall accuracy score of 98% or better for each physician and no critical errors that can adversely affect patient care.

SpeechCheck was

CREATED TO FILL

AN IMPORTANT NEED

in the speech recognition process.

The speech recognition technology vendors do a fantastic job of training physicians on the use of their software, but there needs to be both a consistent method of measuring the quality of final reports and deeper training on the best practices for dictation and self-editing.

The company was started based on the feedback received from facilities who were using speech recognition.

Some of the comments made included the frustration of decreased quality and the fear that one major lawsuit could negatively impact both their bottom line and their reputation.

Technology brings

NEW CHALLENGES

Speech recognition software has been around for a few decades, but has become more widely used over the last 3 to 5 years.

Facilities are looking for ways to decrease expenses, increase efficiencies and improve turnaround times. Speech recognition software accomplishes this by substantially decreasing or even eliminating traditional transcription costs, cutting down clerical staff once responsible for processing final reports and bringing turnaround times down to minutes rather than hours or days. However, the software is not perfect, and there is no way for it to be perfect with the complicated English language and inability to correct physician dictation errors.

The burden of editing these reports

NOW FALLS TO THE

PHYSICIANS,

increasing their workload, decreasing their productivity and increasing the risk of malpractice lawsuits.

SpeechCheck can help lessen that risk by cross-training physicians on how to dictate for speech recognition and how to best edit their reports. SpeechCheck will continue to monitor quality and trends quarterly, establishing a way for each customer to define their standards and measure accuracy on a formal basis.

Physicians spend an average of 22,000 hours to become a physician, but they receive only 30 minutes of training on speech recognition software, often as part of a group and usually just covering the use of the software itself.

The decision to move to speech recognition software, coupled with having physicians self-edit their own reports, is often made at an administrative level within a facility. Physicians have not been trained on making the transition from having a transcriptionist or editor correcting mistakes to being fully responsible for all corrections and the final report.

SpeechCheck will measure the quality of the reports
being sent to patient charts and will assist the physicians with the transition,


HELPING THEM AVOID COMMON

MISTAKES AND PITFALLS.

There are many different types of

SPEECH RECOGNITION ERRORS INCLUDING:

  • Incorrect exam or incorrect patient
  • Findings that do not match impression
  • Grammatical errors
  • Incorrect formatting
  • Incorrect numbers or measurements
  • The Joint Commission compliance issues
  • Left versus right inconsistency within a report
  • Nonsense words or words that are out of context
  • Malpractice suits and litigation
  • Loss of professional reputation
  • Decreased production
  • Decreased referrals from outside physicians, leading to decreased outpatient volume
Request More Information