Health Information Technologies & Processes

Topic: Back-end vs. Front-end transcription quality

1.  Back-end vs. Front-end transcription quality

Posted 11-08-2013 09:00
We are getting ready to move to front-end speech recognition for dictation instead of back-end, but I wonder what kind of quality hit we may see.

With back-end, I am sure that Truman is not the only organization who has doctors who dictate and then when the note shows up for them, they just sign it without reading it.  But, at least they dictate long notes (though costly).

When we go to front-end, and it's all on the docs to edit and review and all of that, I wonder if they will still dictate as thoroughly as they did with back-end, and if they will really take ownership and read their note before signing.

What have others seen?

-------------------------------------------
Seth Katz
Assistant Administrator, Information Governance and Program Execution
-------------------------------------------


2.  RE:Back-end vs. Front-end transcription quality

Posted 11-12-2013 09:26
Seth -
What I have seen is exactly what you fear, at least regarding errors. Much like signing off on transcribed reports, physicians rarely read what is typed. Despite the fact that voice-recognition appears real-time. Products like Dragon improve with use, so you will really need to watch closely when first rolled out. It is also helpful to set up macros/quick-phrases with them when they first start for anything used frequently. E.g. "My normal post-op exam". I have not personally seen a decrease in detail with it though. Only with templates, like PowerNote, have I seen reduction of detail provided.

-------------------------------------------
Susan Clark
EHR Project Manager
-------------------------------------------








3.  RE:Back-end vs. Front-end transcription quality

Posted 11-13-2013 08:53
We're going with Dragon, going to pilot it in the ED and Radiology starting in about two weeks.  I'll keep my fingers crossed

-------------------------------------------
Seth Katz
Assistant Administrator, Information Management and Program Execution
-------------------------------------------








4.  RE: Back-end vs. Front-end transcription quality

Posted 07-13-2014 22:40
Seth - I am curious how implementation of Dragon is going for your organization.  Are you working with a specific vendor?  We are kicking off a new project, starting in Radiology. 

Sandy

-------------------------------------------
Sandy Lewallen
Director, Information Management & Privacy Officer
-------------------------------------------




5.  RE: Back-end vs. Front-end transcription quality

Posted 07-14-2014 08:55
It's going very well.  We sub-licensed it through Cerner, not directly from Nuance, and although the project took way too much time to get up and running, we've seen great clinical adoption, good quality and other providers now proactively asking to use it.

-------------------------------------------
Seth Katz
Assistant Administrator, Information Management and Program Execution
-------------------------------------------




6.  RE: Back-end vs. Front-end transcription quality

Posted 07-16-2014 08:41
How is quality going?  

We see lots of issues related to quality

-------------------------------------------
Krystal Lloyd
Director Health Information
-------------------------------------------




7.  RE: Back-end vs. Front-end transcription quality

Posted 07-17-2014 10:16
It seems to vary by person, really comes down to whether or not they look over their note once it's completed.  Some do a great job, others not so much.  The bigger problem is that some add an addendum that says how this note was made via speech recognition and may have errors.  Our Compliance and Legal teams are reviewing this because that just seems ripe for problems down the road

-------------------------------------------
Seth Katz
Assistant Administrator, Information Management and Program Execution
-------------------------------------------




8.  RE: Back-end vs. Front-end transcription quality

Posted 07-22-2014 17:11

I'm in the same boat now with the speech recoginition errors. I',m in multi specialty group practice. We have providers who include disclaimers such as:
1) as transcribed by voice recognition software
2) documentation elements generated by default template
3) chief complaint and vital signs may be imported from previous visit see intake note
4) " this was done with speech recognition I apologize for any typos/typographical errors
5) Dictated with voice recognition: I apologize for typos + wrong words
6) *THIS NOTE WAS TRANSCRIBED USING VOICE RECOGNITION SOFTWARE*

I took this to our risk manager/compliance officer and referred it to the Chief Medical Officer. It was brought to Clinical Informatics Committee (we do not have a Med Rec Committee). All agreed statements 2 and 3 should not be part of the documentation.!!! Statement 1,4,5,6 was thought to be a non issue "other providers in the communitity do it too"-we want others to know why there are mistakes, and blame the software, lack of time (true point-they are pressured to perform RVUs) etc. and I'm tasked with researching why this is an issue. The CMO was going to search the medical board. I stated if this is blown up in a court room and you have to testify....  Is this a  "I did this note, I didn't proof read it and I don't care"? Risk Management is now researching this for me again - YEAH Ultimately the physician's signature is on the report ...but as Seth said can lead to troubles down the road

Plus not to mention all the errors made and non standard abbreviations the providers are coming up with with completing their own templates. I'm trying to impress that we need Information Governance and compliance. The biggest fear  the CMO has is he doesn't want policies because who would monitor it and enforce it. Sigh.

Thanks!



-------------------------------------------
Kathleen Cleary
Director, HIM
-------------------------------------------




9.  RE: Back-end vs. Front-end transcription quality

Posted 07-23-2014 09:08
We have that issue growing as well and recently discussed it with Legal, Compliance and Medical Staff.  All agree it's a big red flag.  We're doing audits now to track how widespread it is, but may include something in our By Laws that prohibit this.



-------------------------------------------
Seth Katz
Assistant Administrator, Information Management and Program Execution
-------------------------------------------




10.  RE: Back-end vs. Front-end transcription quality

Posted 07-24-2014 10:24
This is a very timely topic.  Since our dictation volume has decreased significantly we are currently using some of our transription staff to perform a montly review (very small sample) of the electronic generated (provider keyed, front-end speech, template) accuracy.  We are looking at discrepancies in a report for example, "no allergies" documented, then there is conflicting statement that says "allergic to penicillin"; abbreviations; copy paste/carry forward; accuracy of typed/front end text.  We are seeing a large error rate.  If we believe it is a signficant error (discrepancy, text does not make sense, etc) we return the document to the provider and ask them to do an addendum to the record.  We are also starting to track by provider. 

I am of the opinion that we might have to elimiate the use of any abbreviations in the electronically inputted documentation because the providers are making up their own abbreviations or using "text messaging" abbreviations, etc.  

 

-------------------------------------------
Marcia Matthias
Corporate Director Health Information/Privacy Officer
-------------------------------------------




11.  RE: Back-end vs. Front-end transcription quality

Posted 07-25-2014 08:57
We struggle with text message talk more on typed nursing notes than anything else.  Because they are so short and to the point, they often use text language.  We haven't added many new abbreviations to our approved list and try to curb shorthand style, but with everything going more mobile and applications like TigerText, I'm sure it'll happen more and more.

-------------------------------------------
Seth Katz
Assistant Administrator, Information Management and Program Execution
-------------------------------------------




12.  RE: Back-end vs. Front-end transcription quality

Posted 9 days ago
Seth,

Are you still using Dragon with Cerner?

------------------------------
Ashley Friedrich
HIM Director
Missouri Delta Medical Center
------------------------------



13.  RE: Back-end vs. Front-end transcription quality

Posted 9 days ago

Yes we are,

 

Seth Jeremy Katz MPH, RHIA
Associate Administrator, Information Management
Truman Medical Centers

Hospital Hill Phone (816) 404-3355 | Lakewood Phone (816) 404-8505 | Fax (816) 404-3142 | Pager (816) 292-9024 | seth.katz@tmcmed.org
Trumed.org |

Description: <a href=">http://www.trumed.org/images/sig_TMC_logo_130.gif">

 

This email message is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message.

 






14.  RE: Back-end vs. Front-end transcription quality

Posted 7 days ago
Seth,

I feel your pain! We are encountering issues with providers failing to proofread adequately before signing their Dragon dictation. I'd recommend looking into the AHIMA/AHDI Clinician Created Documentation QA toolkit. We implemented that here on a small scale. I have two part time transcriptionists that perform the QA and they are very picky! The toolkit contains definitions for critical vs. non-critical errors. We require that all critical errors be corrected. Our coders also send me Dragon errors that I follow-up on. Some of our providers do a terrific job of proofreading but it's that small number that don't that is frustrating. We recently implemented a policy on use of disclaimers. It was reviewed with our attorney. If providers want to use a disclaimer they must use the approved verbiage and not make up their own. They are made aware that the disclaimer does not let them off the hook for the accuracy of their documentation.This is what ours says:

Due to the inherent limitations of voice recognition software, portions of the record may contain occasional wrong word or sound-a-like substitutions. Proofreading has been done to minimize errors. Please call if you need clarification or have questions.

After we went live with Dragon back in 2012 (we are a Cerner site) I did notice some providers didn't have the same level of detail in their reports as they did with transcription. Over time though that seems to have improved. We also offer partial dictations where the provider starts their note, saves it, and then dictates the HPI and impression/plan, transcription inserts it in their note and forwards it on for electronic signature. I wish more would use that process.


------------------------------
Susan Kufahl
Fort HealthCare
------------------------------



15.  RE: Back-end vs. Front-end transcription quality

Posted 4 days ago
Susan,
I'm not finding the toolkit you reference.  Is that in the AHIMA BoK?   If so, is it there by another title?

Thanks,
Marianne

------------------------------
Marianne Dailey
Director HIM & Privacy Officer
------------------------------



16.  RE: Back-end vs. Front-end transcription quality

Posted 3 days ago
Here is the toolkit. I obtained it from AHDI but it was done in collaboration with AHIMA. I don't necessarily use all of the tools attached, but did find the definitions of errors to be very useful. The QA review form is good too. I have implemented that here using the yellow/green highlighting that they show in the sample.

------------------------------
Susan Kufahl
Fort HealthCare
------------------------------