Effectiveness of an AI Scribe Compared to Routine Templates for Clinical Documentation in Orthopedic Consultations

University of Saskatchewan

Status: Completed

Conditions: Clinical Documentation; Artifical Intelligence

Treatments: Other: Routine Dictation Template; Other: AI Medical Scribe

Study type: Interventional

Funder types: Other

Identifiers

NCT07489469

Bio-5718

Details and patient eligibility

About

This randomized study compares AI-generated clinical documentation with traditional dictation templates during orthopedic consultations. Patients are randomized to have their consultation documented using either an AI medical scribe or a routine dictation template. Outcomes include surgeon documentation time, administrative processing time, time from consultation to note delivery to the family physician, patient satisfaction, and documentation accuracy.

Full description

Clinical documentation is a necessary but time-consuming component of surgical practice. Traditional documentation relies on dictation templates that require manual transcription and editing by administrative staff. Artificial intelligence (AI) medical scribes have been developed to automate documentation by recording and transcribing consultations in real time.

The purpose of this randomized study is to compare the effectiveness of an AI scribe versus routine dictation templates in orthopedic consultations.

Patients undergoing consultation for total hip arthroplasty, total knee arthroplasty, or meniscal pathology will be invited to participate. Participants will be randomized in a 1:1 ratio to either AI-generated documentation or standard dictation template documentation. Consultations will otherwise occur according to usual clinical practice.

Data collected will include:

• Surgeon documentation time per patient encounter
• Administrative processing time
• Time from consultation to delivery of the consultation note to the family physician
• Patient satisfaction, including comfort with documentation methods and perceived usefulness of timely note availability
• Documentation accuracy, including spelling, grammar, completeness, and clinical correctness assessed by a blinded reviewer A total of 200 participants will be enrolled. Outcomes will be compared between groups using appropriate statistical tests with a significance level of p < 0.05.

Enrollment: 200 patients

Sex: All

Ages: 18+ years old

Volunteers: No Healthy Volunteers

Inclusion criteria

• Adults attending an orthopaedic consultation for total hip arthroplasty or total knee arthroplasty, and are able to provide informed consent

Exclusion criteria

• Trauma cases or patients undergoing revision surgery

Trial design

Primary purpose: Health Services Research

Allocation: Randomized

Interventional model: Parallel Assignment

Masking: Single Blind

---

200 participants in 2 patient groups

AI Scribe Documentation

Other group

Description:

Orthopaedic consultation documentation generated using an AI medical scribe that records and transcribes the clinical encounter in real time

Treatment:

Other: AI Medical Scribe

Standard Dictation Template Documentation

Other group

Description:

Orthopedic consultation documentation generated using routine surgeon dictation templates with administrative transcription and processing

Treatment:

Other: Routine Dictation Template