Electronic Health Records

Use & Optimization News

Can Natural Language Processing Boost Clinical Documentation?

New research indicates that natural language processing could be helpful in improving clinical documentation, EHR use, and provider workflows.

By Sara Heath

Numerous reports pointed to EHR use as a contributor to increased physician workloads, specifically citing the time demands of clinical documentation processes. According to a recent study, dictation and natural language processing (NLP) may be helpful in reducing these burdens.

“EHR documentation places ever-increasing demands on clinicians’ time, which contributes further to diminished quality of documents (eg, replete with irrelevant, redundant, and erroneous information) and physician dissatisfaction,” wrote lead researcher David R. Kaufman, PhD.

Kaufman and colleagues investigated the use of NLP on time spent on clinical documentation, data quality, and EHR usability. NLP holds considerable potential in clinical documentation improvement, the research team asserted, especially considering physician familiarity with the tools.

The researchers tested four different clinical documentation approaches. First, they tested a purely NLP approach and a purely standard approach using the keyboard and mouse.

The researchers also tested hybrid approaches. The Standard-NLP approach asked physicians to document health history and physician examination data using standard tools and the rest of the visit using NLP. The NLP-Standard approach asked physicians to perform the inverse.

Thirty-one physicians in three specialties – cardiology, nephrology, and neurology – conducted each approach for all of their patients seen. Overall, the physicians demonstrated that clinical documentation approaches incorporating NLP were more effective than pure standard approaches.

The NLP-NLP technique took an average of 5.2 minutes for cardiologists, 7.3 minutes for nephrologists, and 8.5 minutes for neurologists. Standard-Standard took an average of 16.9, 20.7, and 21.2 minutes respectively. The hybrid models both took somewhere in the middle, demonstrating that NLP helps speed up clinical documentation.

The researchers also tested eight data quality measures and ascribed each clinical documentation technique with an average score. NLP-NLP scored a 24.5, Standard-Standard a 29, and Standard-NLP a 29.5.

According to the researchers, the differences in data quality scores were negligible. However, the NLP-NLP technique did score lower potentially because providers were not as precise in reporting health data when they dictated it.

“This suggests that the note was judged to be more to the point and with less redundancy,” Kaufman and colleagues wrote. “In addition, documentation from Standard-Standard Entry was found to be more organized than that from NLP-NLP Entry, indicating that it was structured in a way that the reader could better understand the patient’s clinical course.”

NLP-NLP data entry proved easier to use, receiving a usability score of 36.7. Standard-Standard received a usability score of 30.3.

Overall, the researchers said that NLP proved to be an effective technique for clinical documentation improvement.

“We found that a pure protocol of NLP Entry as well as hybrid protocols (involving both NLP Entry and Standard Entry) showed promise for EHR documentation, relative to Standard Entry alone (Standard-Standard Entry),” the researchers said.

“It is our opinion that different parts of the note should be documented differently, but reaching a conclusion on the optimal method of documentation for each part of the note will require further study.”

These results are particularly important considering the significant role EHRs play in the healthcare industry. Following the 2009 passage of the HITECH Act and the implementation of the EHR Incentive Programs, use of the technology has become widespread.

“This has compelled physicians to adapt to new methods of documentation with concomitant changes to clinical workflow,” the researchers explained. “This has resulted in great uncertainty about the impact of these requirements on the effective application of EHR systems.”

Going forward, healthcare professionals should consider using NLP for clinical documentation. Additionally, researchers should conduct further study to better understand the best approach to implementing NLP in the clinical setting.

“These findings suggest that, pending further study, EHR documentation methods using a combination of dictation and NLP show potential for reducing documentation time and increasing usability while maintaining documentation quality, relative to EHR documentation via standard keyboard-and-mouse entry,” Kaufman and colleagues concluded.

Dig Deeper:




Sign up to continue reading and gain Free Access to all our resources.

Sign up for our free newsletter and join 60,000 of your peers to stay up to date with tips and advice on:

EHR Optimization
EHR Interoperability

White Papers, Webcasts, Featured Articles and Exclusive Interviews

Our privacy policy

no, thanks

Continue to site...