From 3M Health Information Systems
Partial Dictation: A Smart Compromise for Quality Documentation
Last month, AHDI created a new Facebook group called “SR Errors – Funny or Fatal?” as a forum for healthcare documentation specialists (HDS) to share speech recognition “bloopers” that they caught during the editing process.
The submissions vary from hilarious:
“The patient slipped on the ice and fell on her Botox.”
Dictated: “Lipitor 20, two pills a day”
Speech recognition result: “Lipitor 22 pills a day”
The submitters do not share which speech recognition technology they are using or how long the technology has been in place for each dictator, so the errors could be caused by a variety of factors. Also, I want to point out that these issues were discovered via back-end speech recognition (BESR), in which HDS edit the text results behind the scenes and return it to the dictator for final review and approval.
That said, these bloopers make me wonder how much direct-entry EHR documentation is being reviewed and edited in healthcare organizations as the use of dictation and transcription continues to decline. Is there a better approach to documentation capture that is quick and cost-effective but also takes into consideration the need for quality assurance to protect patients, physicians, and the organizations they serve from potentially harmful errors?
For me, the answer is a technology most commonly called “partial dictation” or “partial narrative.” Partial dictation is employed by some EHR vendors to enable physicians to dictate into specific fields or sections in an EHR template so that they can provide more detailed content and context to the patient’s record. For example, the patient’s vital signs, lab values, medications, etc., can be easily captured via discrete data fields in the EHR, but the patient may have a very colorful family or social history that cannot possibly be elaborated via a check box, data field, or pick list. In this example, the physician would be able to select a dictation option for the family history and social history sections, and the resulting dictation “snippets” would be sent immediately to a HDS for processing.
Often, healthcare organizations implement front-end speech recognition (FESR) for narrative sections in the EHR because the dictation becomes text in the EHR in real time and there is no need for transcription. However, physicians are faced with time constraints and distractions while caring for patients, and this environment makes documentation capture challenging. Physicians also typically receive minimal, if any, training, technical support, and editorial assistance. Partial dictation is a good alternative, because the physician can dictate into a specific section and then a HDS can transcribe or edit the resulting text and return it to the EHR for approval.
You may be thinking to yourself, “That sounds great, but transcription is expensive.” Yes, that may be true, but as compared to what, exactly? How much is a physician’s time worth as compared to a HDS? Partial dictation in the EHR is a smart investment because it does eliminate quite a bit of transcription through the use of templates. The dictations are fewer and shorter than the traditional full-document approach, but they contain highly relevant and valuable content that should be subject to a “second set of eyes” to verify accuracy, identify holes or inconsistencies, and draw attention to issues before authentication. Healthcare documentation specialists can even be utilized to flag issues and ask questions of the dictator that may eliminate the need for CDI queries later, which saves both time and money.
I think partial dictation is the happy medium that many organizations need to give their physicians more time for patient care and life, while also providing them with the support they need to document patient encounters accurately and completely. It’s a win-win that every healthcare organization should consider as part of their EHR implementation strategy.
Jill Devrick, product solutions advisor with 3M Health Information Systems, is Immediate Past President of the Association for Healthcare Documentation Integrity (AHDI).