EDITOR’S NOTE: Dr. Erica Remer reported this story live during the January 14 edition of Talk Ten Tuesday. The following is an edited transcript of her reporting.
When I do my Case Western Reserve University documentation course, I sometimes joke about the future, when we won’t need to document because the entire encounter will be videotaped for posterity. Last month, I had to update my “EMR (electronic medical record) is here to stay” talk in response to some articles I saw, and I thought you would find it interesting.
Originally, the EMR presentation included a description of compliant scribe utilization. In July 2017, I did a TalkBack about scribes after a visit to my podiatrist’s office, during which I was informed that some of the providers use a virtual scribe service. I was intrigued by the concept. A remote scribe witnesses the encounter through Google glasses and documents it for the provider in real-time.
Last month, I saw two articles describing a new service of recording an encounter using speech-to-text technology. One described Amazon Transcribe Medical and one detailed Cerner’s offering, but the articles also mentioned that Microsoft and Google have their own versions. I have not seen them first-hand, nor am I endorsing any product.
Subsequently, I attended a funeral at which my husband’s entrepreneurial cousin told me about a business endeavor through which they are recording orthopedist office visits and developing technology to assign codes and bill based on the transcribed encounters. I advised him that the most important element would be for the provider to spell out verbally medical decision-making (MDM) and diagnoses.
I think there is value to verbalizing our clinical findings and detailing our impressions in the presence of the patient. It gives them an understanding of the extent of what we are doing and how much thought goes into our decisions. It gives them the opportunity to share in the decision-making.
Medical decision-making should incorporate the history and the physical examination, the data, and the imaging into the diagnoses and plans. This is how we care for patients. And it is critical to document it, so in case the provider dies in a car accident or wins the lottery and leaves the clinical practice, the next provider can know why they were doing what they were doing, to advance the patient care.
Barring unforeseen changes in the Centers for Medicare & Medicaid Services’ (CMS’s) plans, MDM is also likely to be the factor on which component-based billing is going to hinge. Full disclosure, I support that. I think that the history and physical examination serve the MDM, and that providers get paid for their analysis and synthesis of the information, not how many systems they review or body parts they examine and find normal. But if the provider doesn’t document (or, in this case, verbalize) their thought process, the MDM will be unrecognized and uncredited.
If using encounter-to-chart technology, the clinician will need to read through the transcription for accuracy. Studies show that voice recognition yields an error rate of 7.4 percent, which is decreased by review of the transcriptionist and reduced even further (to a 0.3 percent error rate) when the clinician does a final pass. Providers should never abdicate this responsibility – a missed negative word can change the entire meaning and outcome.
I appreciate and am supportive of wanting to minimize the burden on the documenting provider. However, we must keep in mind that the goal is to take excellent care of the patient. The reason we document is, hopefully, because someone else is going to be reading and acting on what we have recorded. I question how efficient the system can be if the provider doesn’t consider what needs to be in the record, at the expense of the poor end-user having to sift through a lot of extraneous chit-chats and irrelevant verbiage to extract the information.
Finally, I can foresee this degrading the capture of risk-adjusting comorbidities, similar to what implementation of the electronic medical record did. We need to keep “mentation” in the documentation.