Dragon Software

At first, physicians have to train the software on their speech pattern.

Foreign doctors had the most difficult time with this part of the process because sometimes they can't really pronounce a word, and dragon can't recognize it; however, you train 'the dragon' to recognize your speech, but i saw a couple of foreign doctors struggling with this, and as you can deduct, they weren't too happy with the implementation of the software.

Also, many doctors complained they had to slow down their speech so that the software would recognize their 'pattern', and yes, I did see that if they spoke as fast as they did a dictation, the software choked up. So not only do you have to get the speech pattern well, but then you have to speak at a specific speed.

Doctors who dictated fast hated the application, obviously.

I only supported a go live for 4 days with it, so I don't know what the end result was....did the doctors get used to it? Are they liking it better? Did 'metrics' improve because of the dragon software? I wish I knew the answer to those questions and would be nice to hear from facilities who implemented this software to see what their long term experience has been.

My personal observation was that it is definitely not as fast as dictating, but I was only there in the early stages and that's when the 'training the dragon' was taking place. The majority of the doctors hated it, and I wasn't impressed with the software either.

Thanks ikarus. Did they have to speak slower because the Dragon was 'learning' their speech patterns? It sounds like it was probably just that they were always going to have to speak slower.

We had one physician who went and observed another physician use it several years ago, not sure what his opinion of it was at the time though.

Yes, physicians had to speak slower because when they spoke at their "dictation" speed, the software was lagging behind or it simply didn't understand the speech.

Also, there is some room for error. You could be dictating your notes to the dragon and the software, will simply transcribe what it "thinks it heard", not what you said.

For example, you could be dictating, "the patient had a hoorifice cough" and the software might transcribe it as, "the patient had a horse cough."

And that's just a silly example, but the point is that many times I observed there were wrong words on the transcription, and doctors didn't really proofread their dictated notes. Obviously, this wouldn't happen with a human who is transcribing the notes (that type of blatant error at least), and that's why I wasn't really impressed with the software.

The idea is nice, but if I was a doctor, I would definitely be raging if I had to use the software. Then again, this was about 2 years ago, so maybe there has been some improvement.