Although improving access to patient information through electronic health records (EHR) is important, the necessary medical documentation has increased physicians’ clerical load and has been identified as a contributing factor to burnout.
Are there ways to reduce this burden on physicians, and thus decrease burnout, without sacrificing vital data about patient care?
That was the question a task force at The Rothman Institute sought to answer with their analysis of available documentation modalities. Their findings were recently presented at the 2023 annual meeting of the American Academy of Orthopaedic Surgeons.
They evaluated the overall quality of these modalities, including artificial intelligence (AI)-based virtual scribe services, as well as the time needed to capture a patient encounter. Their goal was to maximize physicians’ workload to the fullest extent of their license “by finding methods to outsource certain tasks, such as documentation, as this can be time consuming and redundant,” said study author Michael Rivlin, MD, an orthopaedic surgeon with The Rothman Institute and an associate professor at Thomas Jefferson University, both in Philadelphia.
4 Documentation Modalities
The task force compared the following documentation modalities:
- AI-based virtual scribe service, in which everything said in the exam room is extracted by an AI program running on a tablet
- Medical scribe, in which a person who is either physically in the office visit or participating virtually transcribes the patient encounter
- Transcription service, in which the physician records an audio file about the patient visit and sends it to a third-party company for transcription
- Voice recognition mobile (VRM) application, in which a program available on an EHR platform types the words based on voice recognition
For this prospective study, Use of Artificial Intelligence for Documentation in Orthopaedic Hand Surgery, 3 fellowship-trained orthopaedic hand surgeons evaluated 10 standardized patients with prewritten clinical vignettes. Clinical documentation was performed during the clinical encounter using the AI-based scribe and medical scribe, and then afterwards using a VRM and transcription service.
“Our physicians who were not involved in the documentation acted out these vignettes and each scenario contained an element of distraction to determine if the AI would be thrown off by various nuances that might occur during a clinical visit — such as a parent and a minor sharing their thoughts, or a patient interjecting a story about a friend’s experience with hand surgery in the middle of providing an update on their own surgery,” Dr. Rivlin said.
How They Fared
In total, 118 clinical encounters were documented, including 30 AI scribe, 30 VRM, 28 transcription service, and 30 medical scribe notes. Clinical notes were deemed as acceptable or unacceptable and assigned a letter grade (A, B, C, or F) using an 8-point scoring system. In addition, an attorney reviewed all notes for medical-legal risk.
Overall, all modalities performed well, with similar documentation outputs for each. Specific findings include the following:
- The AI scribe scored significantly lower than the other modalities for 1 specific question: “Is the plan correct?” Although the AI scribe was able to capture most of the verbal and implied elements of medical documentation, formulation of the plan was at times deficient compared with a person acting as scribe. Manual editing of the plan section was required with the AI scribe.
- Documenting clinical encounters through transcription services and VRM applications requires substantial time compared with auto-populated AI-based notes. The average time per note for VRM and the transcription service was 3.48 min and 3.22 min, respectively.
- AI-based scribe services rely on verbalized narrative throughout the entire encounter for accurate documentation, but some verification and correction are needed.
“The AI-based virtual scribe service is a promising tool to help decrease documentation burden without significantly lowering the quality of documentation compared [with] transcription and voice recognition software services,” Dr. Rivlin said.
“[Although] AI has some limitations, it continues to improve as the technology advances. These results create a palette of options for physicians to compare outputs should they want to explore new modalities.”