Transcript Scoring

Document Grading, known as Grading+ is a new tool in the TA+ transcription platform that allows a hospital or medical transcription service organization (MTSO) to grade a single transcript or group of transcripts based on facility-specific standards for quality and accuracy.  With this grading module, transcription supervisors gain efficiency and are able to score many more transcripts in the same period of time compared with using manual methods.  Transcription quality improves as transcriptionists learn from on-line feedback instead of relying on verbal comments or hand-written error notations on hard copies.  TA+ document grading provides on-line scoring to transcription supervisors, MTSOs and individual medical transcriptionists to use for training, evaluation and compensation purposes.  Additionally, TA+ processes the math ‘behind the scenes’ to arrive at a score for the graded document or documents and for the medical transcriptionist or MTSO.  Full reporting of the collected data allows authorized users to analyze improvements in various error categories over periods of time.  The automated display and scoring with Grading+ compliments AHDI (Association for Healthcare Documentation Integrity) recommendations, allowing supervisors to accelerate and standardize transcript grading, a notoriously time-consuming project.

GRADING+ CONSISTS OF THREE PROCESSES

  • Defining error categories and values for the system using an unlimited number of error categories
  • Selecting and grading random or specific transcripts
  • Producing summary quality score reporting

ERROR CATEGORIES AND SCORING METHOD

  • The customer-defined error code is a number assigned to an error category or type
  • Examples of customer-defined error descriptions are ‘Incorrect Punctuation’ or ‘Pertinent Omission’
  • The customer-defined numerical error value is part of the scoring system
  • The quality score is calculated using the following formula: 
    Final Quality Score = 1.0 – (total error value / total lines) * 100
    • For example, using a transcript with 260 lines and an accumulation of 4.5 in error deductions The score would be (1.0-(4.5/260) = .9827)*100 for a score of 98.27
  • Total lines can be configured to include or exclude header and footer lines

SELECTION CRITERIA FOR TRANSCRIPTS TO BE GRADED INCLUDE:

  • Transcriptionist or editor or MTSO
  • Quantity of documents to be graded
  • Date range using either dictation date or transcription date
  • Dictating physician
  • Work type
  • Dictation job number or transcription job number
  • One or more of the above

After the user has manually selected the documents for grading or configured Grading+ to randomly select documents based on criteria provided, the user is presented with a list of documents to be scored.  While reading the text and optionally listening to the physician voice file, graders complete the following steps:

  • Insert the error code at the error location, which Grading+ displays as a superscripted numeral
  • Strike the incorrect text
  • After (to the right of) the strikeouts, replace with corrected phrasing, which appears as blue text

After automatic scoring is completed, graders may:

  • View the computed quality score of the document or group of documents
  • Print the document with error codes, highlighted errors and corrections
  • Print a summary of the errors, by type, and the computed score based on one transcript or several
  • Provide on-line feedback to transcriptionist(s) or MTSO(s)
Search
What's New