Artificial intelligence is proving to be an effective tool for documenting orthopedic encounters in hand surgery

LAS VEGAS, March 7, 2023 /PRNewswire/ — As electronic medical records improve access to patient information, charting increases the office burden for physicians as medical documentation has been identified as a factor contributing to burnout. A new study presented at the 2023 American Academy of Orthopedic Surgeons (AAOS) Annual Meeting analyzed the use of documentation modalities, including artificial intelligence (AI)-based virtual transcription services, to determine overall quality and time required for capture a patient encounter are needed. While AI has shown promise as a tool, accuracy requires some verification and correction.

“In our practice, we have established a task force to better understand and correct physician burnout to examine what we know to be the number one cause of burnout – patient documentation,” said Michael Rivlin, MD, FAAOS, orthopedic surgeon, Rothman Institute and Associate Professor, Thomas Jefferson University, both in Philadelphia. “We wanted to look for ways to maximize the physician’s workload at the maximum level of their license and eliminate stresses that can lead to burnout by finding ways to reduce specific tasks, such as outsourcing documentation, for example, as this can be time-consuming and redundant.”

Because the quality of the AI-generated documentation had not been analyzed, the team compared the quality and time required for documenting common orthopedic encounters in hand surgery using the following modalities during a patient visit:

AI-Based Virtual Typing Service – An AI program runs on a tablet and everything said in the room is extracted by the machine. Medical Recorder – A human who is either physically present at the office visit or attends virtually and writes down the patient encounter. Transcription Service – The doctor uses a voice recorder to record an audio file of the patient’s visit and sends it to a third-party company who transcribes the dictation. Voice Recognition Mobile (VRM) Application – A program available on an electronic medical record platform that inputs words based on voice recognition.

READ :  Legal Considerations for Using Artificial Intelligence in Hiring | Locke Lord LLP

In the prospective study, Use of Artificial Intelligence for Documentation in Orthopedic Hand Surgery, three fellowship-trained orthopedic hand surgeons evaluated 10 standardized patients using pre-designed clinical vignettes. Clinical documentation was performed during the clinical encounter with the AI-based scribe and the medical scribe, and then with a VRM and transcription service.

“Our physicians, who were not involved in the documentation, ran through these vignettes, and each scenario contained an element of distraction to see if the AI ​​would be distracted by various nuances that might arise during a clinical visit – such as: “For example, a parent and a minor sharing their thoughts, or a patient throwing in a story about a friend’s experience with hand surgery while in the middle of reporting on their own surgery,” explained Dr. Rivlin.

A total of 118 clinical encounters were documented, including 30 AI transcriptionists, 30 VRMs, 28 transcription services, and 30 medical transcriptionists. Clinical notes were rated as acceptable or unacceptable and given a letter grade (A, B, C, or F) using an eight-point grading system. In addition, an attorney reviewed all notes for medical legal risks.

Overall, all modalities performed well with similar documentation results. Concrete findings are:

The AI ​​scribe scored significantly lower than the other modalities on a specific question: “Is the plan correct?” – whereas the AI ​​was able to capture most verbal and implicit elements of medical documentation; However, the wording of the plan was sometimes deficient compared to its human counterpart, and manual editing of the plan section was required. Documenting clinical encounters through transcription services and mobile speech recognition applications takes a lot of time compared to auto-filled 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 transcription services rely on verbalized narration throughout the encounter for accurate documentation, but as opposed to a human one Writers need some checking and corrections.

READ :  How Would van Gogh Have Painted the Faroe Islands? | Smart News

“The AI-based virtual transcription service is a promising tool to reduce the documentation effort without significantly reducing the quality of the documentation compared to transcription and speech recognition software services,” said Dr. Rivlin. “Although AI has some limitations, it just keeps getting better as technology advances. These results create a range of options for physicians to compare outcomes when exploring new modalities.”

2023 AAOS Annual Meeting Disclosure Statement

About the AAO
With more than 39,000 members, the American Academy of Orthopedic Surgeons is the world’s largest medical association of specialists in musculoskeletal disorders. The AAOS is the trusted leader in promoting musculoskeletal health. It offers the most comprehensive education of the highest quality to help orthopedic surgeons and allied health professionals at all career levels provide optimal patient care in their day-to-day practice. The AAOS is the resource for information on bone and joint disorders, treatments and related musculoskeletal health care issues; and it is spearheading the healthcare debate about improving quality.

Follow the AAOS on Facebook, Twitter, LinkedIn and Instagram.

SOURCE American Academy of Orthopedic Surgeons