Abstract
BackgroundAutomated methods for identifying clinically relevant new versus redundant information in electronic health record (EHR) clinical notes is useful for clinicians and researchers involved in patient care and clinical research, respectively. We evaluated methods to automatically identify clinically relevant new information in clinical notes, and compared the quantity of redundant information across specialties and clinical settings.MethodsStatistical language models augmented with semantic similarity measures were evaluated as a means to detect and quantify clinically relevant new and redundant information over longitudinal clinical notes for a given patient. A corpus of 591 progress notes over 40 inpatient admissions was annotated for new information longitudinally by physicians to generate a reference standard. Note redundancy between various specialties was evaluated on 71,021 outpatient notes and 64,695 inpatient notes from 500 solid organ transplant patients (April 2015 through August 2015).ResultsOur best method achieved at best performance of 0.87 recall, 0.62 precision, and 0.72 F-measure. Addition of semantic similarity metrics compared to baseline improved recall but otherwise resulted in similar performance. While outpatient and inpatient notes had relatively similar levels of high redundancy (61% and 68%, respectively), redundancy differed by author specialty with mean redundancy of 75%, 66%, 57%, and 55% observed in pediatric, internal medicine, psychiatry and surgical notes, respectively.ConclusionsAutomated techniques with statistical language models for detecting redundant versus clinically relevant new information in clinical notes do not improve with the addition of semantic similarity measures. While levels of redundancy seem relatively similar in the inpatient and ambulatory settings in the Fairview Health Services, clinical note redundancy appears to vary significantly with different medical specialties.
Highlights
Automated methods for identifying clinically relevant new versus redundant information in electronic health record (EHR) clinical notes is useful for clinicians and researchers involved in patient care and clinical research, respectively
We evaluated the incorporation of semantic similarity techniques to our previously described approach based on n-gram language models [10] and compared note redundancy across various specialties and settings
We found that the all three models with semantic similarity measures improve the recall compared with the baseline model, no significant differences in F1measure were found
Summary
Automated methods for identifying clinically relevant new versus redundant information in electronic health record (EHR) clinical notes is useful for clinicians and researchers involved in patient care and clinical research, respectively. One key potential advantage of electronic health record (EHR) adoption is the opportunity for health care organizations to leverage the EHRs for organizational initiatives around improving patient care quality, decreasing costs, and increasing operational efficiency. Copy/paste allows for something that might change over time, like the review of systems (containing the patient’s signs and symptoms) or physical examination, to be directly copied, which potentially results in errors or communication of inaccurate information in subsequent notes [2]. Overall, this leads to long, redundant, and potentially inaccurate EHR notes with irrelevant or obsolete information. A follow-up study demonstrated that applying a visualization cue (highlighting with different color text) to new information as a feature of a prototype user interface saved time for providers reviewing notes and improved note navigation [5]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
