Friday, January 20, 2012

Dragon training, stomping and doing it old school

I have been meaning to discuss this for a while, but tonight seems a good time to do it: Dictation methods. Grossing needs a gross description after all, and it has to get written into the record some way.

In 2010 the practice where I worked made the decision to go from human transcriptionists to Dragon Medical. I got to be the guinea pig for it, followed a week or so later by the other two grossing techs I worked with. We had a miserable few months working out the kinks before they started rolling it out for the pathologists (the transcriptionists at our outlying hospitals were hospital employees not pathology employees so the hospital PAs didn't have to use Dragon). After three or four months (averaging just under a thousand dictations a month per person) it was mostly trained to recognize our speech and we were fairly comfortable using it.

It is trainable, it just takes a while
It does have the advantage that it is easy to create your own customized quick text (aka speed phrases aka macros aka a dozen other things). For instance I set mine up so that when I dictated "start" it typed out the text "A. Received in formalin, labeled "_____"" which saved quite a bit of repetition. And when Dragon is well trained you can speak very fast and it can interpret it, which sometimes meant that just doing a straight dictation was faster than using the text short cuts. It also allows for hands-free dictation, even allowing verbal corrections of things it has incorrectly transcribed. Also, you can see the transcription on the screen in front of you, which allows you to proofread your dictations before they are saved.

There are disadvantages as well. It is a lot of work to get Dragon trained well and some things it refuses to be trained on no matter how many times you try to teach it (every time I dictated the word "yes" it typed out "gas" which made verbally navigating through CoPath's menus ridiculously difficult). Dragon requires you to dictate everything and only transcribes what you dictate, which can be frustrating. A good transcriptionist is great to work with, they know what you should have in a dictation for common specimen types and can fill in the blanks when things are inadvertently left out. Dragon doesn't do any of that. The start up cost can be high and you really have to have a computer screen at each grossing station.

Also, in cases where you have people on rotations (any academic or training facility) the time required to train Dragon to a useable level is longer than the time the residents or students would be spending on the rotation. That is also the sort of situation where you want to have the well trained ear of a practiced transcriptionist proofing anything they dictate.

Before Dragon Medical came along we had human transcriptionists and foot pedals, similar to the set up at Duke. I loved this set up! It allowed for hands free dictation, so like Dragon, you could be finishing your dictation while sectioning/filtering/closing up the cassettes. It was efficient. The disadvantage was that if you had to go back and make a correction you had to rewind and listen to your dictation to see what you had already said, which wasn't bad on a one line GI dictation and awful on a three minute long kidney dictation. And, there is the recurring salary cost of transcriptionists that you don't get with Dragon.

And before Dragon and before foot pedals there are the old school methods. Before my old company got foot pedals we had hand held voice recorders. It also relied on transcriptionists so it had the advantages of having a live person going over the dictation and didn't require a computer to be dedicated to each grossing station. It was clunky, did not allow for hands free dictation, and the dictation tapes had to be physically delivered to our transcriptionists (who worked from home). It did seem to rewind faster than the foot pedals though, which made doing corrections easier. There is also the option of typing or writing our your dictation but I've never really used those methods so your idea of the pros and cons is probably as valid as mine is.


  1. I've never used Dragon Medical, but I used to work with a woman who tried to use Dragon software for her office work. It was hilarious (and eventually very annoying) to listen to her repeat the same words over and over again to try to get the software to recognize it. I distinctly recall her getting so frustrated once that she repeated the word "f^(%" over and over again - I don't think the software got "f^(%" right a single time.

  2. I've never used any voice-recognition software, and I don't care to. If I'm forced to use such a program because that's all a particular pathology department has, then obviously I will have to use it. But coming from the "P.A. School of Hard Knocks," I've always used the foot pedal-microcassette method (or the foot pedal-digital method). And to me, the foot pedal method (FPM) is bar-none the best out of them all.

    With FPM, the disadvantages that you described above occur very rarely, and there are ways to work around that. If you're good with dictations, there should be no reason to have to go back and edit a nephroblastoma. If, by some chance, you DO have to go back, what I used to do is write out the sentence(s) that I needed and then gave that to the transcriptionist with the original requisitions so that she could make the necessary edits. Either that, or I'll leave her a note saying "please see me when you get up to dictation number etc. etc." This remedial method would save you not only time, but also the hassle of having to dictate all over again.

    As far as the recurring salary costs of transcriptionists, that's really of no worry to hospitals in the least (unless they're looking to cut costs severely and drastically). The biggest fear with voice-recognition dictation software is that there would be no need for transcriptionists anymore. You could speak, out comes the dictation, and then it gets printed or faxed over to the pathologist. But in this economy where everyone is talking about "JOBS, JOBS, JOBS", it may not be wise to get rid of medical transcriptionists. I contend that they're actually healthy for the economy (if only by a minute amount).

    Lastly, the hand-held voice recorders were a real pain at this one job I had! The biggest disadvantage with those is getting the thing all bloody, which could then damage the equipment. I had no choice but to keep my gloves on while dissecting a placenta! I certainly was not going to change gloves every time I had to dictate a sentence just to keep the recorder clean. It was a bloody mess! Having to change the batteries every now and then was also a huge nuisance. Figures how that job lasted only a month, and I certainly would never put that on a resume....It was the most disorganized, disgusting, and dysfunctional pathology department I've ever set foot in. Even the pathologists themselves were incompetent, and the Chinese supervisor I had (who had difficulty speaking English) really got on my nerves a lot. I was actually about to be fired, but I quit before they could terminate me. Whenever I think about this one hospital, my blood starts boiling because of how ridiculous it was. But I hate to go on another rant again, so if you want to know exactly why I was going to be fired, let me know and I'll tell you.

  3. And speaking of old school, you should write a blog on cassette and slide labeling. I always used to write the specimen accession numbers on the cassettes in pencil or tissue-marking pens. I'm not sure how advanced most hospitals are these days, but where I come from, the hospitals all used very antiquated equipment.

    I first saw the beautiful Leica cassette printer at a private lab job I had right after graduation. In fact, I had to write on the cassettes when I first got hired, but then they gradually incorporated the cassette printer into the job. And boy, was it neat! I loved that thing! All the specimen bottles had bar codes on them, so you just scanned the bar codes into the computer, typed in how many cassettes you needed in whatever sequence, and the printer did the rest.

    All this old-school stuff makes me think: As George Carlin once said, "You know why my immune system is so good? I swam in RAW SEWAGE!" From a Pathologist's Assistant perspective with my experiences, I can totally see what he means lol.

    1. It sounds like you could write about it since you have had exposure to a wider variety of labs/hospitals than I have. :)

    2. I may have to, but I'll do that when the time comes, I guess. An experience on the job would have to occur for me to write passionately about a topic like that. It's gotta be an "in the moment" kind of thing haha