Loading ...

TOPLINE:

Using an AI-guided VALUE framework (Validate, Align & Reframe, Link & Educate, Unite in a plan) framework for preoperative consultations about metabolic and bariatric surgery (MBS) improved decision-making in residents over those who used self-directed learning.

METHODOLOGY:

• Outcomes after MBS depend heavily on the quality of the preoperative consultation, which is critical for patient engagement and adherence. Surgical residents often lack structured training, and AI could help scaffold the development of communication skills.
• Researchers conducted a randomized trial involving 40 surgical residents (postgraduate years 1-5) to assess the VALUE framework, an AI-guided metacognitive tool, on shared decision-making and communication outcomes in MBS consultations.
• Participants were assigned to an AI-guided group (n = 20; mean age, 30.8 years; 85% male), which used the VALUE framework with a large language model, or a self-learning control group (n = 20; mean age, 31.1 years; 90% male), which prepared consultations without generative AI; each resident conducted two 15-minute, video-recorded simulated consultations with two different standardized patients.
• In the AI-guided group, residents had 15 minutes per case to use the VALUE prompt to create a personalized consultation outline, which they could use as a thinking aid but not as a script, and they could not refer back to it during the simulated consultation.
• The primary outcome was the score on the nine-item Shared Decision-Making Questionnaire (SDM-Q-9; 0-100 scale), rated by standardized patients immediately after each encounter; higher scores reflected greater shared decision-making.

TAKEAWAY:

• The AI-guided group achieved a mean SDM-Q-9 score of 84.7, whereas the self-learning group achieved a mean score of 71.3, yielding a mean difference of 13.4 (P < .01). This difference exceeded the minimal clinically important difference of 10 points, confirming clinical relevance.
• Residents in the AI-guided group had significantly lower scores on the Decisional Conflict Scale than those in the self-learning group (mean difference, -12.6 points; P < .01), with the reduction in scores exceeding the minimal clinically important difference of 10 points.
• Blinded observers rated AI-guided consultations higher on the Four Habits Coding Scheme (mean difference, 2.7 points; P < .01), with notable improvements in eliciting perspectives, demonstrating empathy, and investing in the end.
• The AI-guided group also reported greater increases in communication self-efficacy and produced higher-quality preparatory consultation outlines.

IN PRACTICE:

“For clinical practice, the VALUE framework offers a structured, efficient method to transform a routine preoperative consult into a values clarification and alignment session,” the authors of the study wrote.

“Our findings suggest that the AI-guided VALUE framework is a promising hypothesis-generating strategy for improving shared decision-making in MBS consultations. By scaffolding the clinician’s preparatory reasoning, the intervention produced encouraging preliminary improvements in patient-perceived collaboration, decisional conflict, observed communication skills, and surgeon self-efficacy,” they added.

SOURCE:

The study was led by Chun Gao, Tongji Hospital, Wuhan, China. It was published online in Obesity Surgery.

LIMITATIONS:

The trial had a small sample size. The observed effect sizes were unusually large for a single 15-minute structured preparation session. The study tested one specific AI model and one specific communication framework, so it is not yet clear whether the same results would occur with other models or frameworks.

DISCLOSURES:

The study does not list specific funding sources, and the authors declared having no competing interests.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.