Real-time Assessment of Dialogue in Motivational Interviewing Training (ReadMI)
Project Final Report (PDF, 339.07 KB) Disclaimer
Disclaimer
Disclaimer details
A training solution using artificial intelligence can be tailored to allow providers to quickly and actively address inadequate training, and thus improving their ability to elicit from patients their own motivations to make healthy behavior changes.
Project Details -
Completed
-
Grant NumberR21 HS026548
-
Funding Mechanism(s)
-
AHRQ Funded Amount$293,128
-
Principal Investigator(s)
-
Organization
-
LocationDaytonOhio
-
Project Dates08/01/2019 - 07/31/2022
-
Population
-
Type of Care
-
Health Care Theme
Motivational interviewing (MI) is an evidence-based, patient-centered approach to interaction with patients focused on eliciting from patients their own motivations to make healthy behavior changes. However, MI can be challenging for providers to use due to inadequate training and the tendency for physicians to take on a more active role in patient care.
In this study, researchers developed a software-based training tool, ReadMI, to enhance the skill development of doctors and other healthcare workers in the use of MI with patients by analyzing clinician responses and providing them with instant feedback on how to enhance MI abilities.
The specific aims were as follows:
- Develop the ReadMI training tool with speech-recognition software that provides MI training metrics with 95 percent accuracy.
- Test the 95 percent accurate ReadMI in a randomized controlled trial.
Researchers evaluated the effectiveness of a tool called ReadMI for improving MI skills in medical students and residents. The study consisted of two parts: in the first, the ReadMI tool was refined, and its accuracy in analyzing MI skills was evaluated using transcripts from 48 interviews conducted by medical residents with a simulated patient. In the second part, a randomized controlled trial (RCT) was conducted with 125 third-year medical students to determine the benefits of adding ReadMI feedback to MI training. The results of the RCT showed that the intervention group, which received feedback using the ReadMI tool, had significantly better MI skills compared to the control group.
The research showed that artificial intelligence can be used to measure and improve skills related to MI in a time-efficient manner. ReadMI provided real-time quantitative feedback to learners during MI training and reduced the cognitive load on the training facilitator. The use of ReadMI was found to be advantageous in improving specific communication skills and valuable in measuring and quantifying changes in performance. The tool is intended to contribute to the wider use of MI and improve the ability of health professionals to effectively engage and activate patients in chronic disease prevention and management.
Disclaimer
Disclaimer details