Free EMR Newsletter Want to receive the latest news on EMR, Meaningful Use, ARRA and Healthcare IT sent straight to your email? Join thousands of healthcare pros who subscribe to EMR and EHR for FREE!

AI-Based Tech Could Speed Patient Documentation Process

Posted on August 27, 2018 I Written By

Anne Zieger is veteran healthcare consultant and analyst with 20 years of industry experience. Zieger formerly served as editor-in-chief of FierceHealthcare.com and her commentaries have appeared in dozens of international business publications, including Forbes, Business Week and Information Week. She has also contributed content to hundreds of healthcare and health IT organizations, including several Fortune 500 companies. Contact her at @ziegerhealth on Twitter or visit her site at Zieger Healthcare.

A researcher with a Google AI team, Google Brain, has published a paper describing how AI could help physicians complete patient documentation more quickly. The author, software engineer Peter Lui, contends that AI technology can speed up patient documentation considerably by predicting its content.

On my initial reading of the paper, it wasn’t clear to me what advantage this has over pre-filling templates or even allowing physicians to cut-and-paste text from previous patient encounters. Still, judge for yourself as I outline what author Liu has to say, and by all means, check out the write-up.

In its introduction, the paper notes that physicians spend a great deal of time and energy entering patient notes into EHRs, a process which is not only taxing but also demoralizing for many physicians. Choosing from just one of countless data points underscoring this conclusion, Liu cites a 2016 study noting that physicians spend almost 2 hours of administrative work for every hour of patient contact.

However, it might be possible to reduce the number of hours doctors spend on this dreary task. Google Brain has been working on technologies which can speed up the process of documentation, including a new medical language modeling approach. Liu and his colleagues are also looking at how to represent an EHR’s mix of structured and unstructured text data.

The net of all of this? Google Brain has been able to create a set of systems which, by drawing on previous patient records can predict most of the content a physician will use next time they see that patient.

The heart of this effort is the MIMIC-III dataset, which contains the de-identified electronic health records of 39,597 patients from the ICU of a large tertiary care hospital. The dataset includes patient demographic data, medications, lab results, and notes written by providers. The system includes AI capabilities which are “trained” to predict the text physicians will use in their latest patient note.

In addition to making predictions, the Google Brain AI seems to have been able to pick out some forms of errors in existing notes, including patient ages and drug names, as well as providing autocorrect options for corrupted words.

By way of caveats, the paper warns that the research used only data generated within 24 hours of the current note content. Liu points out that while this may be a wide enough range of information for ICU notes, as things happen fast there, it would be better to draw on data representing larger windows of time for non-ICU patients. In addition, Liu concedes that it won’t always be possible to predict the content of notes even if the system has absorbed all existing documentation.

However, none of these problems are insurmountable, and Liu understandably describes these results as “encouraging,” but that’s also a way of conceding that this is only an experimental conclusion. In other words, these predictive capabilities are not a done deal by any means. That being said, it seems likely that his approach could be valuable.

I am left with at least one question, though. If the Google Brain technology can predict physician notes with great fidelity, how does that differ than having the physician cut-and-paste previous notes on their own?  I may be missing something here, because I’m not a software engineer, but I’d still like to know how these predictions improve on existing workarounds.

EMR Note Cloning is Scarier than I Thought

Posted on June 15, 2012 I Written By

Dr. Michael J. Koriwchak received his medical degree from Duke University School of Medicine in 1988. He completed both his Internship in General Surgery and Residency in Otolaryngology-Head and Neck Surgery at Vanderbilt University Medical Center. Dr. Koriwchak continued at Vanderbilt for a fellowship in Laryngology and Care of the Professional Voice. He is board certified by the American Board of Otolaryngology-Head and Neck Surgery. After training Dr. Koriwchak moved to Atlanta in 1995 to become one of the original physicians in Ear, Nose and Throat of Georgia. He has built a thriving practice in Laryngology, Care of the Professional Voice, Thyroid/Parathyroid Surgery, Endoscopic Sinus Surgery and General Otolaryngology. A singer himself, many of his patients are people who depend on their voice for their careers, including some well-known entertainers. Dr. Koriwchak has also performed thousands of thyroid, parathyroid and head and neck cancer operations. Dr. Koriwchak has been working with information technology since 1977. While an undergraduate at Bucknell University he taught a computer-programming course. In medical school he wrote his own software for his laboratory research. In the 1990’s he adapted generic forms software to create one the first electronic prescription applications. Soon afterward he wrote his own chart note templates using visual BASIC script. In 2003 he became the physician champion for ENT of Georgia’s EMR implementation project. This included not only design and implementation strategy but also writing code. In 2008 the EMR implementation earned the e-Technology award from the Medical Association of Georgia. With 7 years EMR experience, 18 years in private medical practice and over 35 years of IT experience, Dr. Koriwchak seeks opportunities to merge the information technology and medical communities, bringing information technology to health care.

The health IT community is well aware of the dangers of cloning notes in an electronic medical record.  I include myself in that group.  Until recently I prided myself for doing a good job, both in our EMR design and in my own personal practice, of using just the right amount of automation in our documentation workflow.  Two recent events showed me that I still have some work to do.

The first event occurred a few weeks ago when I was reviewing some records.  One patient note documented an enlarged salivary gland containing a stone.  That would be fine except for one small detail – I had removed that gland one week prior to the date of the note!  My nurse had created that note.  A conversation with her revealed she thought she was doing the right thing by always clicking the “previous finding” button, which I had programmed myself.  My nurse is extremely bright; this was my fault for not training her on this issue.  I had also signed that note.  So it was my fault twice.  After a 30 second conversation with my nurse it has not happened since.

The second event was when an attorney interviewed me regarding one of my patients.  I was a treating physician in a malpractice case (I am not the defendant thankfully).  The attorney wanted to know if, in my opinion, the physician defendant had met the standard of care in treating the patient despite the adverse outcome.

This was a high-risk case for note cloning; the patient had multiple abnormal neurologic findings that were stable over time.  In reviewing my records I was satisfied that my notes were accurate, complete and original for every visit.  I avoided cloning those abnormal but stable findings by describing the same exam but using slightly different wording at each visit.  How else do you avoid cloning?  But the attorney pounced on my small changes in description, trying to establish a trend in my notes that the patient was getting worse.  I explained the cloning issue to him, and he understood…. I think.  Nonetheless I felt somewhat uncomfortable defending my documentation, and I was not even the defendant.  In trying to avoid cloning notes I had stepped right into another problem.

This issue is huge in my practice.  I have a large volume of head and neck cancer patients.  The essence of caring for them properly is to monitor them for changes in their abnormal – but stable – physical findings.  A recurrence of cancer might manifest as a subtle change in one of these findings.

How do you document that an examination is stable and unchanging, but change your wording enough to document that you actually examined the patient at every visit?  We do not yet have the cloning issue figured out.