Artificial intelligence as good as other methods for documenting hand surgery cases
Click Here to Manage Email Alerts
LAS VEGAS — Automatic-populated clinical notes that are artificial intelligence-based may decrease documentation burden for hand surgery cases in the future vs. transcription services or voice recognition systems, according to a presenter.
In research presented at the American Academy of Orthopaedic Surgeons Annual Meeting, Michael Rivlin, MD, FAAOS, discussed results of a prospective study he and his colleagues conducted to compare the quality of documentation of common orthopedic encounters for patients who undergo hand surgery with an artificial intelligence (AI)-based virtual scribe service, a transcription service and a voice recognition mobile (VRM) application.
Image: Adobe Stock
Provider time is a critical resource that needs to be conserved, particularly with the high rates of physician mental health issue and burnout today, Rivlin told Healio in an interview.
“The more provider time you can give back to the provider, the more they can spend on caring for patients, so modalities, such as AI scribe or an in-person scribe, are helpful, but over time we expect the AI scribe to replace the human factor,” Rivlin said.
Because the quality of AI-generated documentation had not been analyzed previously, Rivlin and colleagues compared the quality and time spent documenting common orthopedic encounters in hand surgery, according to a press release. They used the following modalities during a patient visit:
- AI-based virtual scribe service, which is an AI program that runs on a tablet and a machine extracts everything said in the room;
- medical scribe, which a person either physically in the office visit or who participates virtually and transcribes the entire patient encounter;
- transcription service, in which the physician uses a Dictaphone device to record an audio file about the patient visit and sends it to a third-party company so that what has been dictated can be transcribed; and
- VRM application, which is a program available on electronic medical record platforms that types the words said based on voice recognition.
According to a press release for the study, three fellowship-trained orthopedic hand surgeons evaluated 10 standardized patients with pre-written clinical vignettes. Physicians not involved in the study acted out the clinical vignettes, which were documented by the different systems being analyzed, according to the press release.
Clinical documentation was performed during the clinical encounter using the AI-based scribe and medical scribe, and then afterward using a VRM and transcription service.
According to the press release, a total of 118 clinical encounters were documented. This included 30 AI scribe, 30 VRM, 28 transcription service and 30 medical scribe notes. Clinical notes were deemed as either acceptable or unacceptable and assigned a letter grade of A, B, C or F using an eight-point scoring system. In addition, an attorney reviewed all notes for medical legal risk.
Overall, all modalities performed well with similar documentation outputs between the modalities. According to the abstract, there were no significant differences in note quality for each hand surgeon who participated in the study, as well as for the entire cohort.
“I do expect AI will continue to pop up in more and more physician offices across the country,” Rivlin said.
“The AI will take the clinical burden off our hands,” he told Healio.
Collapse
Bridges T, et al. Poster 170. Presented at: American Academy of Orthopaedic Surgeons Annual Meeting; March 7-11, 2023; Las Vegas.
Click Here to Manage Email Alerts