Let me start with what Ms. Johnston got at least partly right in her blog post:

There are incredible numbers of veteran MTs who are now crying because they are seeing their paychecks fall precipitously because there’s no way to double production to make up for half the wages.

MT wages are stagnant or falling and have been for some time. Are employers asking medical transcription speech recognition editors to work for less per line? Yes, they are. But let’s get over the victim mentality here and take some responsibility for the pay situation we’re all in. The reason MT pay rates aren’t better is because there’s a never-ending supply of people who are willing to work for less, whether they’re new to the field or whether they’re veterans. I’ve encountered very few MTs who have said “I’m not willing to work for that” – and meant it.

The fact is, there are MTs who are accepting less than half their production typing line rate and that’s where they start getting into problems.

The math does work out – if you are 2x more productive editing speech and your line rate is half what it was for manual transcription, then you are making the same amount of money. And there are MTs who are making production 2x and more of their manual typing production.

If you produce 200 lines/hour on average as a transcriptionist at 8 cpl your average hour rate of pay is $16/hour

If you produce 400 lines/hour on average as a speech editor at 4 cpl, your average hourly rate of pay is $16/hour

The problem starts if you are not more productive and if you accept a line rate that is much less than your production increase. A lot of speech recognition jobs are paying 2 and 3 cpl. C’mon people – this isn’t rocket science. You’re all smart enough to do the math. Assuming that you will actually BE 2x more productive as a speech editor, you can’t take a line rate that is less than half your current rate if you want to break even!

The other problem in this equation is the problem that has plagued medical transcription almost from the get-go: it’s a Pink Collar Ghetto job, which means it’s predominated by female workers. And studies have shown that women do not negotiate their pay like men do.

Some companies allow their MTs to trial as speech editors. If you try it and you aren’t making enough money – either go back to traditional transcription or renegotiate your pay. Some companies are paying a full line rate on jobs that require XX% retyping, acknowledging that for some jobs generated by a speech recognition engine, the accuracy is so poor that it basically has to be completely retyped.

Now let me get to what’s inaccurate and what bothers me about this blog post.

I’m not sure how Ms. Johnston can state with such utmost certainty the conditions and attitude in the industry when people who have more experience and more connection with what’s going on in the industry couldn’t make these statements with any confidence in their accuracy. It’s as if Ms. Johnston has her finger on a thready extremity pulse and is telling the everyone, including the doctor, who is monitoring the EKG, that she’s certain the patient is dying.

There are incredible numbers of veteran MTs who are now crying because they are seeing their paychecks fall precipitously because there’s no way to double production to make up for half the wages.

How does Ms. Johnston knows that there’s “an incredible number” of anybody crying over falling pay rates due to SR? I’m quite certain there’s an incredible number of veteran MTs who aren’t happy about pay rates in ANY segment of transcription, but SR hasn’t penetrated a sufficient segment of the market to impact “an incredible number” of MTs with any number of years of experience. See – I can state things with absolute certainty, as well – it doesn’t make them facts, nor does it make them true.

This statement also ignores the opportunity SR provides to veteran MTs who are embracing speech recognition editing because they are no longer capable of doing the production work required to make a living. They do far less keyboarding as editors and their knowledge continues to be utilized. It’s a welcome option for veterans who don’t want to have to learn something completely new (like, for example, cancer registry).

…no one’s ever bothered to ask for our input on the front end of this abomination.

Holding onto that thready extremity pulse, the author makes this statement like she actually knows what she’s talking about. The fact is, the medical transcription industry IS heavily involved. It was a topic in at least two sessions given by medical transcription industry professionals at a meeting I recently attended. You know all those things you say AHDI doesn’t do for transcription? Well, this is one of the things they are doing and they’ve been involved in it for years. What would be accurate is that nobody has bothered asking the author’s input – but then again, she hasn’t positioned herself to give it where it counts, either. The transcription industry acknowledges that speech editing requires a slightly different skill set than manual transcription, but the same knowledge base. Transitioning MTs to that skill set is part of the focus of AHDI as they attempt to move MTs to what is becoming the future of transcription. The fact that a lot of MTs don’t know this is part of AHDI’s perception problem with MTs. And if your idea of AHDI “doing something” about SR is to get them to make it stop, you need a bigger reality check than I can give you.

Adding to my amusement at this assertion is a recent article published by Advance for Health Information Professionals, which is the sponsor of Ms. Johnston’s Passage blog. Maybe Ms. Johnston doesn’t read Advance; maybe she just missed this article: “Creating a Definitive Guide on Speech Recognition.”

The automated speech recognition technology (ASRT) work group, which consists of a broad group representing speech recognition vendors, medical transcription service organizations (MTSOs), MTs and consumers, hopes to clear up the confusion that exists by publishing what the group hopes will be a definitive and evolving guide to speech recognition, in the form of a reference guide to the adoption of speech recognition.

MTs who want objective information about SR and the medical transcription industry should follow the link and read the entire article. If you’re really interested in making a difference, join the workgroup.

Like Ms. Johnston, there are MTs who see SR as a “Borg assimilation,” rather than an opportunity. And like Ms. Johnston, they may very well find other careers in healthcare information management. We saw MTs kick and scream and cry about transitioning out of DOS and WordPerfect 5.1 and into Windows. We all survived and adjusted and by golly – some of us even like it better. And now we have many MTs who wouldn’t know a DOS prompt if it smacked them upside the head. Resistance to change is a human attribute. In my experience, the personality type attracted to MT is more resistant to change than most others. But get real folks – you can’t stop progress. At some point, you have to decide whether you’re going to lead, follow or get out of the way. I’d like to think we all have the intelligence to be rational and put emotions aside.

We all get a good guffaw out of speech recognition errors, but I’ll bet MTs are less entertained when their own errors are posted or distributed in a company newsletter. If an MT gets 90% accuracy on a QA score and speech recognition gets 95% – which one would you pick? The fact that MT speech editors are seeing reports with errors doesn’t equate to the blanket statement “speech recognition doesn’t work.” If an MT consistently produces reports with 99% accuracy, is her work sent to QA? Of course not. If an MT consistently produces reports with 90% accuracy, is her work sent to QA? Of course it is. The same is true of documents produced by speech recognition – once a consistently accurate report is generated for a dictator, those reports are no longer sent to editing. Keep in mind that M*Modal, eScription and others like them aren’t in business because the technology doesn’t work. HIMs managers do talk to one another and they know salesmen are there to sell a product – the technology has to prove itself, and it has.

For MTs to be successful in making the transition to speech recognition – and make money – they need real information, not misinformation and not har-har jokes about errors.

If you are asked to transition to speech recognition, you need to know what questions to ask so you can negotiate your pay. And you do need to negotiate the conditions and pay.

• Is there a sliding scale pay rate based on the accuracy percentage of a report? In other words, do I get paid more for a report that is only 85% to 94% accurate, versus my base rate for reports that are 95% accurate or higher?
• If accuracy falls below 75% and I have to retype most of the report, do I get full transcription rate pay?
• What is the average percent accuracy of speech recognition on the account(s) I’ll be working on?
• How long has the account been on speech recognition? (The length of time will impact the accuracy – newer accounts will have a lower rate of accuracy.)