Epic Research shows note bloat increased by 8%
Photo by: JGI/Tom Grill/Blend Images
A dual team study of notes written by 166,318 outpatient providers in the U.S. from May 2020 to April 2023 published by Epic Research on Thursday examined how coding requirements and documentation tools have influenced clinical documentation length.
WHY IT MATTERS
The 2021 changes to the Centers for Medicare and Medicaid Services' evaluation and management CPT billing codes aimed to reduce the administrative documentation burden on providers, Epic Research said in a new report.
To understand if the changes had an effect, the researchers evaluated 1.7 billion clinical notes in electronic health records written by 166,318 outpatient providers across primary care and specialties.
They found the providers averaged 4,628 characters per clinical note in 2020 and increased to an average of 5,002 characters in 2023 (8%).
However, providers generally reduced their average time of 5.4 minutes spent on each note to 4.8 minutes per note during the same period.
While overall average note length increased during the study period, approximately 40% of providers did reduce their note length average. A full 10% of those providers spanned primary care, internal medicine, surgical specialties, dermatology, cardiology, psychiatry and other specialties.
"This suggests that a reduction in note length is achievable in nearly any specialty," researchers said.
"Furthermore, nearly 90% of providers reduced the average time they spent writing each note."
Looking at composition methods, the Epic researchers also determined that healthcare organizations that reduced their use of SmartTools – documentation tools that make it easy to add additional content to notes from other places – and minimized copying and pasting within the patient chart also reduced their average note length.
THE LARGER TREND
Last year, researchers from the University of Pennsylvania Perelman School of Medicine in Philadelphia published a JAMA study about their work using artificial intelligence to analyze all UPenn Health System notes over a five-year period ending in 2020.
Natural language processing found a preponderance of duplication across all the notes for 1.96 million unique patients. Half of the words were duplicated from prior notes, and the longer the record, the greater the degree of duplication.
This earlier study concluded the time-based and author-based organization of modern EHRs drive the prevalence of note duplication.
"Duplicate text casts doubt on the veracity of all information in the medical record, making it difficult to find and verify information in day-to-day clinical work," the researchers said.
Recently, Epic and Nuance announced new clinical documentation features combining conversational and ambient AI with GPT4 to turn provider and patient conversational interactions into clinical documentation providers can edit and approve.
"This collaboration will allow our physicians to focus more completely on the care and treatment of their patients while the AI works behind the scenes to document the encounter, allowing people and technology to each do what they do best without delays," said Dr. Hal Baker, CDIO at WellSpan Health, in an Epic statement about the new application.
ON THE RECORD
"Organizations that increased note length saw stable use of SmartTools and increased use of copy/paste functions," the researchers said in the report.
"These findings align with previous research that found increased use of SmartTools and copy/paste functions were correlated with longer notes."
Andrea Fox is senior editor of Healthcare IT News.
Email: afox@himss.org
Healthcare IT News is a HIMSS Media publication.