On The Revival Path - Traditional Transcription Industry

Today, there are digital recorders with voice recognition software for dictation, mind-disk transcription systems, digital conference transcription and digital interview transcription. Web-cast and pod-cast transcription using digital recorders with non-destructive editing features and various record modes are also available in the market. The HITECH Act and ‘meaningful use’ have further driven healthcare organizations to generate efficiencies to a great extent both in terms of return on investment and patient care. 

Traditional Transcription to Stay

While speech recognition practices are gaining momentum, a recent study by the American Journal of Roentgenologysuggests the need to have a re-look at the concerns voiced in the past about the potential for errors in this new technology. According to this study as reported in the For–the-record-magazine website, reports generated using automatic speech recognition system were likely to contain more errors than those generated with traditional dictation transcription. The study centered around reports on complex cases of breast imaging using both automatic speech recognition and conventional dictation transcription. Anabel Scaranelo, PhD, MD, of Toronto-based University Health Network and coauthor of this study says that “at least one major error was found in 23% of automatic speech recognition reports compared to 4% of conventional dictation transcription reports.” It also found a wide difference in the rate of error when these breast MRI reports were examined separately. While the error rate for speech recognition reports stood at 35% that for conventional reports it was only 7%. 

Modern Pathology, co-authored by Walter Henricks MD, Cleveland Clinic and published in 2002, did not find favor with the voice recognition technology for its limited utility and cost inefficiencies. Out of the two speech recognition methods used in a healthcare setting i.e. front end and back end speech recognition, experts like Henricks believed that front end work flow strategies are injudicious because the clinician dictates into the speech recognition engine as well as edits the document for ensuring accuracy. The editing job is a distraction as “Pathologists need to be doing pathology work,” Henricks says. He apprehends that in speech recognition errors cannot be easily identified because the technology always uses a real word and it would be difficult to catch the true error with the naked eye. 

However, experts like him believe that transitioning transcriptionists into the role of editors could become a staple of speech recognition deployment. They also think that there can be a reduction in the Transcription services but it cannot be totally done away with, as electronic systems begin to generate elements such as continuity-of-care documents. 

Moving Ahead

Speech recognition software has several advantages in comparison with other methods but the question before the clinician is how to calculate the risks against the benefits. This is where the traditional transcription needs to score over. However, it needs to be remembered that the future of Medical transcriptionindustry depends on its willingness and ability to transform transcription services in accordance with the changes in the medical environment.